วีดีโอ

This approach to lung nodule calcification patterns was taught to us when I was resident, and continues to be what my colleagues and I continue to teach our residents today. It appears in textbooks we assign for reading, and in the scientific literature too. Take for example, the chest radiology textbook "Diagnostic Thoracic Imaging" by Wallace T. Miller (a radiology textbook favored at Penn): "Amorphous, irregular, punctate, and eccentric patterns of calcification have been identified in a variety of malignancies including bronchogenic carcinoma, carcinoid tumors, and metastasis." "It must be remembered that calcification alone is not diagnostic of a granuloma or a benign condition. Calcification will be present in 6% to 14% of primary lung carcinomas. However, the calcification in cancer is typically amorphous or stippled in character, different than the patterns of calcification which are specific for granulomas." Similar discussions of the different lung nodule calcification patterns and their implications go back for decades in the scientific literature. Take for example the 1993 AJR paper "CT of the Lung: Patterns of Calcification and Other High-Attenuation Abnormalities" by Chai and Patz: "Approximately 6% of all primary lung cancers show a punctate, amorphous, or reticular pattern of calcification on CT scans. This variation is probably due to several different causes: (1) engulfment of benign calcification by the tumor as is seen in scar carcinoma, (2) dystrophic calcification arising from necrosis within the tumor, and (3) calcium deposition resulting from secretions by the tumor."

"Infiltration" is a term that's not commonly encountered with respect to the CT scans. If folks do use this term, it's typically on chest x-rays when they see a nonspecific lung opacity. Since CT imaging usually affords us the capability to be more specific in characterizing a lung opacity, the need to use a "catch-all" nonspecific term like "infiltration" is much less. Many subspecialist chest radiologists - myself included - discourage the use of the term "infiltration" altogether - not be cause it's nonspecific, but because it means different things to different people and is therefore ambiguous in its meaning. For some folks, an "infiltration" could represent atelectasis, infection, non-infectious inflammation, hemorrhage, neoplasm, or interstitial fibrosis in the lung, while for other folks it might represent a subset of these items, and for some it might just mean "probably pneumonia". Since we strive to avoid miscommunication that may affect clinical management, a term like "opacity" is favored since it tends to have a more consistent interpretation by all parties.

In their 2017 article "Recommendations for Measuring Pulmonary Nodules at CT: A Statement from the Fleischner Society", the Fleischner Society recommends: "the dimension of small pulmonary nodules (<10 mm) should be expressed as the average of maximal long-axis and perpendicular maximal short-axis measurements in the same plane. For larger nodules and masses, both long- and short-axis measurements should be recorded." The reason for their recommendation is probably because the solid lung nodule size thresholds that appear in the current follow-up CT recommendation tables published by the Fleischner Society (for incidental pulmonary nodules) and the American College of Radiology (for lung cancer screening) currently are *average* diameters, rather than short-axis or long-axis diameters. For example, a footnote at the bottom of the Fleischner Society 2017 Guidelines for Management of Incidentally Detected Pulmonary Nodules in Adults reads: "Dimensions are average of long and short axes, rounded to the nearest millimeter."

You can find a link to a PDF of all of these slides, just by going to the video and expanding the “Description” section underneath. You’ll find links of this type for almost every one of my talks in the Description area of each video.

Thank you! I’ve had a really tough time finding a good book or paper on this topic, and have had to piece things together along the way during my career. That’s why I felt driven to make this particular talk!

@@radiologyframeworks its unfortunate that there is not a resource. This topic is NEVER talked about in residency---you better believe the CPA's and MBA's doling out the physician $$ at hospitals can recite this information.

It’s really rewarding to know they’re helpful!

Thanks! You'll encounter a range of performance reported throughout the medical literature, and the reference I used was one that seemed closest to my personal observations over the years. The reference is: "CT and diffusion-weighted MR imaging in randomized order: diffusion-weighted imaging results in higher accuracy and lower interrater variability in the diagnosis of hyperacute ischemic stroke" by Fiebach et al. in Stroke (2002). A reference that reported lower sensitivity, but higher specificity is "Certainty of Stroke Diagnosis: Incremental Benefit with CT Perfusion over Non-contrast CT and CT Angiography" by Hopyan et al. in Radiology (2010), where a sensitivity of 52.5% and specificity of 84.4% are reported for radiologists diagnosing stroke on non-contrast head CT. There's another reference that reports even lower sensitivity and higher specificity for non-contrast head CT: "Magnetic resonance imaging and computed tomography in emergency assessment of patients with suspected acute stroke: a prospective comparison" by Chalela et al. in The Lancet (2007), where they report: For acute stroke: - CT sensitivity 26% - MR sensitivity 83% - CT specificity 98% - MR specificity 97% For acute ischemic stroke: - CT sensitivity 16% - MR sensitivity 83% - CT specificity 98% - MR specificity 96%

Glad you liked it!

Thank you for the great question. In order to answer, I will need to borrow some talking points from my advanced version of this talk (the version I've uploaded here on TH-cam is the basic version). 1. I mentioned there is a superoinferior blood perfusion gradient that favors the lower lungs. There is ALSO a radial blood perfusion gradient that favors the central lungs over the peripheral lungs. Pulmonary vessels in peripheral lung are, on average, smaller in caliber than those in the central lung, which means that pulmonary vascular resistance is higher in peripheral lung than central lung. Higher pulmonary vascular resistance results in slower vascular transit time. Slower transit times are why we see greater deposition of things like septic emboli and metastases in the peripheral vascular bed of the lung than its central vascular bed. 2. There is a FOURTH factor that influences the geographic distribution of diseases in the lungs, besides blood perfusion, lymphatic flow, and ventilation. This fourth factor is mechanical stress. Mechanical stress is greatest in the upper lung and the peripheral lung. Imagine you're hanging off the ledge of a cliff by just your hands, and someone below you is hanging on to your ankles... that's sort of like how the upper lung feels. The upper lung "carries" the weight of the lower lung, which results in more connective tissue strain and alveolar distention in the upper lungs. Surface tension keeping the visceral and parietal pleural surfaces adherent to each together result in more connective tissue strain in the peripheral lungs. UIP/IPF have a BASAL predominance because: (1) higher blood perfusion in the lower lungs results in greater deposition of substances that may contribute to the development of UIP/IPF (e.g. chemotherapy agents, inflammatory mediators, etc.) and (2) alveoli are smaller in the lower lungs at end-expiration, which means higher surface tension inside the alveolar sac, greater possibility of alveolar collapse, and subsequently traction on adjacent bronchioles. UIP/IPF have a PERIPHERAL predominance because: (1) mechanical stress is higher in the peripheral lungs, and (2) vascular transit time is slower, resulting in greater deposition of substances that may contribute to the development of UIP/IPF. Hope that helps!

@@radiologyframeworks Thank you very much, were could we find the advanced version of this lecture?

Glad it was helpful!

Although I imagine it could happen on occasion on a chest x-ray, it would be uncommon with CT imaging.

You're welcome. Let me see what I can put together!

Gosh, you're right. Thanks for catching that! It seems the Apple Keynote movie export (that I uploaded here in TH-cam) shows that cine video clip differently than when I present in person. I'll upload an image under the "Community" tab for this channel to show the MFH example better.

You're welcome! Keep an eye out for a brain MR focused talk, coming very soon…