1	Dangerous Areas in Head CT Interpretation : Lessons Learned From A Long-term Resident Quality Assurance Process. Vagal AS^1,3, Leach JL^1,2,3, Strub WM¹, Tomsick TA^1,3, Lukin RR^1,3 Departments of Radiology University Of Cincinnati College of Medicine¹ Cincinnati Children’s Hospital Medical Center² and The Neuroscience Institute³ Cincinnati, Ohio
2	Disclosures The authors have no financial disclosures .
3	Objectives We have a robust quality assurance for resident preliminary interpretations at our institution. This serves as a large database, from which common regions and scenarios of misinterpretations are illustrated. This exhibit will highlight the areas of Head CT where subtle pathology may be difficult to detect. A rational approach to these “danger areas” is presented with recommendations for identification along with multiple case examples.
4	“The hardest thing to do is call an imaging study negative” Anthony A Mancuso ,MD “The hardest thing we do is call a chest radiograph negative” Leo Rigler, MD circa 1974
5
6	Introduction Head CT is typically first line imaging test in patients with acute neurologic symptoms. Interpretations may be challenging, especially in the acute setting when abnormal findings may be subtle. A robust quality assurance process can assist in documenting misinterpretations and form the basis of resident education efforts with the goal of improving patient care.
7	Methods QA data from 20,234 CT scans of the head was collected from January 2004 to October 2007 on patients at University of Cincinnati Medical Center. These head CT scans were interpreted by radiology residents on overnight call from 5.00 pm to 7.00 am at an academic level I trauma center. The patient population included trauma and non-trauma patients seen in the emergency department as well as hospital inpatients. The CT scans were then reviewed the following morning by staff neuroradiologists, all with CAQ in Neuroradiology.
8	Methods Discrepancies were documented in the final dictation along with communication with the emergency department or the ordering in house physician. Patient charts were reviewed for imaging follow up and clinical outcomes. Discrepancies were divided into major and minor categories. Major category includes a misinterpretation which potentially delayed clinical management and could have resulted in mortality or morbidity. Minor discrepancies were defined as not impacting immediate care of the patient.
9	Results
10	Results
11	Discussion
12	Misinterpretations by On- Call radiology Residents Multiple studies have looked at the accuracy of interpretations by radiology residents as compared to the staff radiologists (1-4). All these studies have shown a low rate of discrepancies in the resident preliminary interpretations . Also the potential of adverse clinical outcome is rare. Erly et al found a 2% major and a 7% minor discrepancy rate on 1324 preliminary head CT examinations. Major discrepancies included vasogenic edema misinterpreted as ischemia, missed suprasellar mass , acute infarcts and contusion.(4)
13	Misinterpretations by On- Call radiology Residents Wyoski et al found a 1.7% major and a 2.6% minor discrepancy rate on 419 preliminary head CT examinations. Major discrepancies included subarachnoid, acute infarcts and contusion.(3) Lal et al found a 0.9% rate of significant misinterpretation in 2388 neuroradiological scans.head CT examinations.(2) Strub et al, from our group, found 0.1% major and a 3.2% minor discrepancy rate in 5206 cranial and spinal CT studies (1)
14	Misinterpretations by On- Call radiology Residents The purpose of this exhibit is to identify common areas of Head CT misinterpretation in order to reduce discrepancies, improve patient care, and more effectively use this knowledge for resident education. Multiple examples of discrepant preliminary resident head CT studies are illustrated in the following slides. The misinterpretations are broadly categorized into missed hemorrhage , mass effect /edema, missed masses and missed ischemia.
15	Intracranial hemorrhage
16	"A previous large study of..." A previous large study of resident head CT interpretations from our center focused on the patterns of missed intracranial hemorrhage .(5) Similar patterns of misidentified hemorrhage are noted when evaluating these cases as a group.
17
18
19
20
21
22
23
24
25
26
27	"Key points for detecting subtle..." Key points for detecting subtle hemorrhage : Subtle areas of hidden blood include interpeduncular cistern , sylvian fissures, dependent hemorrhage in the occipital horns and region of foramen magnum. Commonly missed subdural hematomas are in midline and along the tentorium , likely because of the attenuated appearance of the falx and the tentorium.(5) Small contusions are easily missed along the inferior temporal and frontal lobes due to the partial volume averaging of the adjacent bone. Maximise the window settings (approximately W 110-115 , L 65-75) to look for small hemorrhages. Also sagittal and coronal reconstructions may be helpful.
28	Missed fractures
29
30
31
32
33	Key points for missed fractures The majority of the major discrepancies in this category in our study were skull base fractures. In a trauma patient, especially with multiple other major findings, it is easy to overlook the subtle skull base fractures. An airfluid level or opacification in the sphenoid sinus should prompt a very careful search for a fracture, with special attention to the carotid canal. The same detailed search should be made for a temporal bone fracture in cases of mastoid or external ear opacification. Always use the thin section bone algorithm for this purpose. Reconstructed datasets (1-2 mm sections) are invaluable in this setting.
34	Missed mass effect and edema
35
36
37
38
39
40
41
42
43
44
45
46	Key points in missed edema /mass effect The importance of looking at the fourth ventricle in every head CT case cannot be stressed enough.The fourth ventricle should be visible in at least one section of the posterior fossa and should be in the midline. Mass effect and effacement of this ventricle may be the only clue of the abnormality. The region of the foramen magnum can be a blind spot for masses, aneurysms, hemorrhage and low lying tonsils. In suspicious cases, sagittal and coronal reconstructions may be very helpful. The cortical sulcal symmetry should be part of the checklist. Any asymmetry needs to be explained –either due to patient head tilt or mass effect. The recognition of difference in the pattern of vasogenic edema and cytotoxic edema is important.
47	Missed Ischemia
48
49
50
51
52
53	Key points for missed ischemia Use of a narrow CT window and level setting (approximately window width 18-21 and level 27-30) can increase the sensitivity for detection of early ischemic changes.(6). This accentuates the gray-white matter differentiation and can detect early edema. Observation of CT eye deviation improves the identification of acute ischemia, especially in the absence of specific clinical information as to the side of the stroke. (7)
54	Missed masses
55
56
57
58
59
60
61
62
63
64	Missed Sinus thrombosis
65
66	Extracranial Findings
67
68	Resident education It is important to identify repetitive patterns of misinterpretation for educational purposes. At our institution, discrepancies from overnight cases are collected and shown to residents quarterly as a part of quality assurance lecture series. The common areas of misses are highlighted by attendings in the daily neuroradiology checkouts, especially to the junior level residents. A test is given to the first year residents, using these cases, before having them participate in overnight call.
69
70	“Structured checklist for Head CT interpretation” Bones (including skull base evaluation) Pay attention to sinus and mastoid opacification as a secondary sign of skull base and temporal bone fracture in trauma patients. Dedicated windowing, reconstruction algorithm essential. Extracranial soft tissues Orbits Ventricles and Subarachnoid spaces includes checking for sulcal symmetry and position and caliber of the fourth ventricle. Brain parenchyma: Assess parenchymal attenuation at different window settings to maximize pathology detection.
71	“Danger areas checklist ”as identified by our database Foramen magnum blind spot for blood , masses, aneurysms and low lying tonsils. Fourth ventricle Size and position Mass lesions or aneurysms in the sellar and suprasellar region Dural sinuses Check these areas for hidden hemorrhage: Interpeduncular cistern Sylvian fissures Occipital horns Check these areas for subdural / epidural hemorrhage Tentorium Parafalcine (asymmetric density along the falx is abnormal) Along convexities (use altered window settings to detect hemorrhage along the calvarium)
72	Conclusion Head CT is the most commonly performed examination in Neuroradiology and misinterpretations are not uncommon. Application of the knowledge gained from a long term robust QA process can elucidate these “ Danger areas”, improve interpretation accuracy, and ultimately improve patient care.
73
74	References 1. Strub WM, Vagal AA, Tomsick T,Moulton JS. Overnight resident preliminary interpretations on CT examinations: should the process continue? Emerg Radiol 2006;13:19–23. 2 .Lal NR, Murray UM, Eldevik OP,Desmond JS. Clinical consequences of misinterpretations of neuroradiologic CT scans by on-call radiology residents. AJNR Am J Neuroradiol 2000;21:124–29. 3. Wyoski MG, Nassar CJ, Koenigsberg RA, Novelline RA, Faro SH, Faerber EN.Head trauma: CT scan interpretation by radiology residents versus staff radiologists. Radiology 1998;208:125–28. 4. Erly WK, Berger WG, Krupinski E,Seeger JF, Guisto JA. Radiology resident evaluation of head CT scan orders in the emergency department. AJNR 2002;23:103–07. 5. Strub WM, Leach JL, Tomsick T, Vagal A.Overnight preliminary head CT interpretations provided by residents: locations of misidentified intracranial hemorrhage. AJNR 2007 Oct;28(9):1679-82. 6. Lev MH, Farkas J, Gemmete JJ, Hossain ST, Hunter GJ, Koroshetz WJ and Gonzalez RG.Acute stroke: improved nonenhanced CT detection—benefits of soft-copy interpretation by using variable window width and center level settings. Radiology 1999 213: 150–155. 7. Mahajan,V Minshew PT, KhouryJ, Shu PP, Muzaffar M, Abruzzo T, Leach JL, and Tomsick TA. Eye Position Information on CT Increases the Identification of Acute Ischemic Hypoattenuation. AJNR 2008 Mar 20 [Epub ahead of print]. 8. Leach JL, Fortuna RB, Jones BV, and Gaskill-Shipley MF. Imaging of Cerebral Venous Thrombosis: Current Techniques, Spectrum of Findings, and Diagnostic Pitfalls RadioGraphics 2006 26: S19-41S.