1998 Joseph E. Whitley, MD, Award. Relationship between American College of Radiology in-training examination scores and American Board of Radiology written examination scores. Part 2. Multi-institutional study.

RATIONALE AND OBJECTIVES: The authors assessed the relationship between resident performance on the American College of Radiology in-training examination and subsequent American Board of Radiology written examination. MATERIALS AND METHODS: Percentile scores from the in-training examination (low score, < 20 percentile) and written board examination (low score, < 25 percentile) were collected for 513 residents from 11 university-based programs over a 7-year period. Mean in-training examination scores were compared for the high- and low-score written board examination groups. In-training examination scores were correlated with the written board examination scores. An odds ratio was calculated for the association between in-training examination and written board examination scores; this was adjusted for resident Alpha Omega Alpha (AOA) status, clinical training prior to radiology residency, the year the written board examination was taken, program size, and regional location of the program. RESULTS: The mean in-training examination scores were significantly higher in the high-score than the low-score written board examination group (P = .0001). There was a significant correlation between the in-training examination score and the written board examination score (P = .0001). The crude all-high in-training scores-written board scores odds ratio was 9.618, and the adjusted all-high odds ratio was 7.595. The final model included resident AOA status. CONCLUSION: Resident average in-training examination score was a strong predictor of the written board examination score, as noted in the earlier report. The resident with a low in-training examination score is at risk for poor performance on the written board examination.

Relationship between American College of Radiology in-training examination scores and American Board of Radiology written examination scores.

RATIONALE AND OBJECTIVES: To retrospectively evaluate the association between resident performance on the American College of Radiology in-training examination and performance on the American Board of Radiology written examination. METHODS: Percentile scores from the in-training examination (low score, < 20th percentile) and written board examination (low score, < 25 percentile) were collected for 58 residents in a large, university-based program during a 6-year period. Mean in-training examination scores were compared for the high score and low score written board groups. In-training examination scores were correlated with the written board scores, and odds ratios were calculated for the association between in-training examination and written board scores. Adjusted in-training examination and written board odds ratios were calculated for Alpha Omega Alpha status and prior clinical training. RESULTS: The mean in-training examination scores were statistically significantly higher in the high score written board group (P = .0001). There was significant correlation between the in-training examination and the written board scores (P = .05). There was a significant association between a resident's average in-training examination score and written board score. Alpha Omega Alpha status was associated with high written board scores, and prior clinical training was associated with low written board scores (not significant). CONCLUSION: The resident's average in-training examination score was a strong predictor of the written board score. The resident with a low in-training examination score is at risk for poor performance on the written board examination and may benefit from remedial training.

Postmyelography headache: a review.

The author reviews the topic of postdural puncture headache (PDPH) so radiologists can better understand, prevent, and treat this common complication of myelography. PDPH, a postural headache relieved by lying supine, occurs after 36.0%-58.2% of myelographic procedures performed with 22-gauge Quincke needles. It begins within 3 days and lasts 3-5 days. Thin, young women are more likely to develop PDPH. Smaller gauge needles, blunt-tip needles, a paramedian approach, and parallel orientation of a bevel needle are associated with lower PDPH rates. Intravenous hydration, no removal of cerebrospinal fluid, experience of the myelographer, and inpatient status have been inconsistently associated with lower PDPH rates. Treatment is supportive with bed rest. Oral or intravenous methylxanthine agents may be given for more severe PDPH. An epidural blood patch, although rarely needed, is very effective.

Postmyelography headache rates with Whitacre versus Quincke 22-gauge spinal needles.

PURPOSE: To compare the postdural puncture headache (PDPH) rates after myelography with use of 22-gauge Whitacre (blunt tip) versus standard 22-gauge Quincke (bevel-tip) spinal needles. MATERIALS AND METHODS: At myelography, 340 patients were randomly assigned to one of two groups in which either the Whitacre (167 patients) or the Quincke (173 patients) needle was used. Follow-up was at 48 hours and at 4 days after the procedure. Crude and adjusted Whitacre/PDPH odds ratios and risk ratios were calculated. RESULTS: The crude Whitacre needle PDPH risk was 9.6%, and the crude Quincke needle PDPH risk was 15.6%. The absolute risk difference was 6.0% with a 95% confidence interval of -0.98% and 13.04%. The adjusted Whitacre/PDPH odds ratio was 0.492 with a 95% confidence interval of 0.241 and 1.003. The Whitacre needle group had a statistically significantly lower PDPH severity grade (P = .0151), similar PDPH duration, and more technical difficulty with the needle. CONCLUSION: The Whitacre needle is associated with an equal or lower PDPH risk; the lack of statistical significance is most likely related to sample size. Further study is necessary to determine whether the decreased PDPH rate justifies the added difficulty associated with use of the Whitacre needle.

Evaluation of 1H magnetic resonance spectroscopic imaging as a diagnostic tool for the lateralization of epileptogenic seizure foci.

The purpose of this study was to assess whether a visual examination of 1H spectroscopic images could correctly lateralize patients with intractable temporal lobe epilepsy. 20 patients with intractable temporal lobe epilepsy and 10 volunteers were included in this study. Spectroscopic images were analysed using a protocol based on visual inspection. Images of the metabolites N-acetyl aspartate (NAA), choline (Cho), creatine (Cr) and lactate were obtained from a transverse plane oriented along the sylvian fissure. Images from each individual were evaluated independently by six reviewers. Results of the lateralization procedure obtained from the visual examinations were compared with those obtained from quantitative analysis of the spectra and with those obtained by magnetic resonance imaging (MRI), positron emission tomography (PET), neuropsychological examinations, and electroencephalographic (EEG) recordings. NAA images were found to be the most effective, amongst metabolite images, in lateralizing the epileptogenic lobe. Using the site selected for resection as the definition of the correct lateralization, 70% of the patients who underwent temporal lobectomy were correctly lateralized by the majority of the examiners using the visual inspection protocol. Based on the results of this study it is concluded that visual examination of 1H spectroscopic images is potentially valid in lateralizing patients with intractable temporal lobe seizures. Confidence in the visual interpretation increased as the difference in NAA signal intensity between the temporal lobes increased. The threshold above which the majority of the examiners correctly lateralized the patients was approximately 15% in NAA signal loss in the ipsilateral lobe.

The paranasal sinuses of patients with acute graft-versus-host disease.

PURPOSE: To determine whether patients who have acute graft-versus-host disease (AGVHD), are more likely to have radiographically abnormal paranasal sinuses after bone marrow transplantation than patients without AGVHD. METHODS: This retrospective case-control study population comprised 45 adult allogeneic or matched unrelated donor patients who underwent bone marrow transplantation. Twenty-eight patients had AGVHD, and 17 patients did not. All patients had paranasal sinus imaging with either CT or plain films for evaluation of possible sinusitis after bone marrow transplantation. The severity of radiographic changes was measured with an objective scoring system based on mucosal findings. RESULTS: An odds ratio revealed no association between AGVHD and the presence of radiographically abnormal sinuses. CONCLUSION: The diagnosis of AGVHD should not complicate the radiographic evaluation of acute sinusitis after bone marrow transplantation.

Design of clinical radiologic research.

The components of clinical research design--the research question, the review of prior research, the study design, the study subjects, the study measurements, the statistical analysis, and the critical analysis--have been discussed. These components relate to the overall philosophy of the research process in the following manner. The driving force behind research is the desire to understand a universal truth (Fig. 6). The research question is formed to reflect the truth sought. Next defined is the specific target population and the phenomena of interest with regard to the sought-after truth. The study is then designed with a study plan detailing timing and location and an intended sample population chosen to reflect the target population. The phenomena of interest are translated into measurable intended variables. Care is needed to identify the systematic error that can be introduced at this stage; outside opinion can be invaluable. The actual study results from the implementation of the study design. The actual subjects may differ from the intended sample population due to a low recruitment rate or a large dropout rate. The actual measurements may differ from the intended variables due to unsuspected confounding variables or random errors. Statistical analysis is performed on the study findings to infer truth within the study. This truth may be affected by random and systematic errors in implementing the study. Truth in the study is then inferred to reflect truth in the universe. Systematic and random errors in study design may affect the validity of that deduction, but that deduction is the only way we have of understanding universal truth.(ABSTRACT TRUNCATED AT 250 WORDS)

Neurologic complications of petrosal sinus sampling.

Petrosal sinus sampling has been described as an innocuous, essentially risk-free procedure. The authors report on two patients from two different institutions who sustained brain stem injury as a result of petrosal sinus sampling and two other patients in whom brain stem injury was avoided. Major neurologic complications have occurred in only 0.2% of procedures (one of 508) performed at the National Institutes of Health and in 0.5% of a subset of these procedures (one of 184) that were performed with a specially designed petrosal sinus catheter. The cause of these complications is unclear but is presumed to be localized venous hypertension. Tip-deflector catheter-guide-wire systems, if available, are recommended for petrosal sinus sampling. Inadequate data exist to permit recommendation of any other catheter. Brain stem injury is preventable if the catheter is withdrawn at the earliest sign of even a minor, seemingly insignificant problem. Subtle symptoms and signs that may not appear to be neurologic may herald a clinical catastrophe if not heeded.

Endoscopic paranasal sinus surgery: radiographic evaluation of severe complications.

PURPOSE: To report our experience with the radiographic evaluation of severe complications resulting from the functional endoscopic sinus surgery (FESS) procedure. PATIENTS: Ten major complications were reviewed retrospectively. FINDINGS: Ten major complications occurred. Eight of 10 had injury to the floor of the anterior cranial fossa, fovea ethmoidalis (roof of the ethmoid sinus), or roof of the sphenoid sinus. Six patients presented with meningitis or rhinorrhea, two presented with headache and massive pneumocephalus; one patient who presented with meningitis had a large nasal frontal encephalocele. Noncontrast brain CT that included the paranasal sinuses adequately evaluated the source of pneumocephalus. Thin-section coronal CT accurately predicted the site of leak in five patients. Both coronal sinus CT and MR imaging were useful to confirm the nasal encephalocele. Two of 10 had vascular injury secondary to FESS. One patient presented with subarachnoid hemorrhage seen on noncontrast CT and cerebral angiography demonstrated an aneurysm of the anterior cerebral artery. The second patient suffered severe intraoperative hemorrhage. Emergency angiography revealed a pseudoaneurysm of the cavernous carotid artery, and balloon occlusion of the artery was performed. No deaths occurred in this series. CONCLUSION: Radiologists should be familiar with the rare, but potential complications of this commonly performed procedure in order to help direct the work-up in an efficacious manner.

Improved detection of cerebral hypoperfusion with internal carotid balloon test occlusion and 99mTc-HMPAO cerebral perfusion SPECT imaging.

A new method that employs a cerebral perfusion imaging agent, 99mTc-HMPAO, is described for identifying those patients who clinically pass an internal carotid artery balloon test occlusion but who then may develop a cerebral infarction after permanent occlusion. Test balloon occlusion of the internal carotid artery was performed in 17 patients, and 99mTc-HMPAO was injected intravenously while the balloon was inflated. The balloon was deflated and removed approximately 15-20 min later, and single-photon emission CT (SPECT) imaging of the brain was performed. In patients with an area of focal hypoperfusion, the 99mTc-HMPAO study was repeated the next day to determine if the hypoperfusion was due to the temporary balloon occlusion or to a preexisting abnormality. All 17 patients had normal neurologic examinations at the time of the balloon test occlusion. Fifteen had symmetric perfusion on the 99mTc-HMPAO SPECT scans. Five of these had permanent internal carotid artery occlusion, and four did not develop a neurologic deficit. One patient developed an ipsilateral watershed infarction after internal carotid artery occlusion during significant intraoperative hypotension. Two patients had ipsilateral regions of hypoperfusion with the test occlusion. Both these patients had repeat nonocclusion 99mTc-HMPAO SPECT studies that were normal. One patient is being followed. The second patient had ipsilateral EEG changes associated with temporary intraoperative occlusion of the internal carotid artery during aneurysm surgery. Although the numbers are relatively small, this study suggests that this protocol may be able to predict a good outcome after permanent carotid artery occlusion as well as to identify patients who are at greater risk for developing cerebral infarction after permanent occlusion.(ABSTRACT TRUNCATED AT 250 WORDS)

Pituitary adenoma: correlation of half-dose gadolinium-enhanced MR imaging with surgical findings in 26 patients.

Sellar magnetic resonance imaging studies obtained with half doses of gadopentetate dimeglumine (0.05 mmol/kg) were prospectively interpreted and retrospectively rated in 26 patients who subsequently underwent transsphenoidal sellar surgery for suspected pituitary adenoma. Studies included a sagittal scout view followed by a non-contrast-material-enhanced, an immediate postcontrast, and a delayed postcontrast T1-weighted image (obtained at 1.0 or 1.5 T). Ten of 11 confirmed microadenomas were identified prospectively; all were identifiable in retrospect. Macroadenomas (12 cases) were well demonstrated. The high signal intensity of the posterior pituitary and of intrasellar hemorrhage was obscured on postcontrast studies. Delayed images proved unnecessary. This prospective evaluation suggests that a half-dose study is comparable to retrospective studies in which full-dose techniques were used for detection of micro- and macroadenomas. Imaging times are reasonably short, and cost of contrast material is potentially reduced. Confirmation with larger studies is required, and careful endocrinologic and clinical follow-up of nonsurgical patients is necessary.

Magnetic resonance artifact in the postoperative cervical spine. A potential pitfall.

An magnetic resonance imaging artifact that stimulates hypertrophic bone formation is described in patients who have had an anterior cervical discectomy. The magnetic resonance images of 26 patients with anterior cervical discectomy were retrospectively reviewed. Comparison was made to the available concurrent computed tomographic scans, computed tomographic myelograms, and operative reports. A bovine spine was drilled with a drill only at one level and with a metal suction tip in close proximity to the drill at another level, and magnetic resonance images were obtained. Artifact was present in 12 patients and absent in 14; this was confirmed in the 8 patients with comparison studies. Close correlation was seen with the prospective reading of the presence of artifact and operative drill use in the seven patients with available operative reports. The bovine spine model showed no artifact at the drill-only level and significant artifact at the level where the metal suction tip was positioned next to the drill. Small metal flecks were seen grossly at the second level, but not on plain roentgenograms. The metallic magnetic resonance artifact seen in postoperative cervical spines is probably from small bits of metal from the metal suction tip as it occasionally hits the drill. Bone abnormalities seen on magnetic resonance imaging at the level of a previous anterior cervical discectomy may need a cervical computed tomogram to confirm the findings.

Diagnosis of cerebral metastases: double-dose delayed CT vs contrast-enhanced MR imaging.

For patients suspected of having cerebral metastases, double-dose delayed CT (DDD-CT) has proved significantly more sensitive than CT scans obtained immediately after administration of a lesser dose of iodinated contrast material. Previous reports confirm the advantages of postcontrast MR imaging over contrast-enhanced CT, but data comparing DDD-CT and contrast-enhanced MR have not been reported. This study describes comparative imaging results in 23 patients who had contrast-enhanced MR imaging to clarify equivocal findings on DDD-CT studies. Contrast-enhanced MR demonstrated more than 67 definite or typical parenchymal metastases. T2-weighted MR revealed more than 40, while DDD-CT revealed only 37 typical metastatic lesions. Three patients had five or fewer lesions on DDD-CT and lesions "too numerous to count" on MR. The frequency of equivocal or unconvincing lesions was similar on DDD-CT (11) and contrast-enhanced MR (10). On T2-weighted images, we noted a substantially higher number of equivocal lesions (19), fewer definite metastases, and a number of definite metastases that had no corresponding lesion on the enhanced studies, confirming the inability of T2-weighted imaging to specifically identify cerebral metastases. In one case, multiple tiny lesions on T2-weighted images were not apparent on DDD-CT scans and were recognized only in retrospect on contrast-enhanced MR images. In this series, MR with enhancement proved superior to DDD-CT for lesion detection, anatomic localization of lesions, and differentiation of solitary vs multiple lesions. Cost-benefit considerations precluded a comparison between the two techniques in all patients suspected of having cerebral metastases.(ABSTRACT TRUNCATED AT 250 WORDS)

Nuclear magnetic resonance imaging (NMR), (MRI), of brain stem tumours.

The NMR scans of twenty-six patients who were referred with the clinically suspected or histologically verified diagnoses of brain stem tumour were reviewed. Twenty-one patients (eleven children and ten adults) had scans that were positive for tumour. Nine of these had histological confirmation. Eleven were irradiated without histology on the basis of the clinical diagnosis. The remaining patient was treated expectantly. The other five patients (two children and three adults) had CT and NMR scans that were negative for tumours. Clinical follow-up of these five cases for 9 to 29 months disclosed no further evidence of tumour. All but one tumour had evidence of an increased T1 and T2. Mass effects were seen in all but one case (not the same as the first exception). Multiplanar imaging was useful in determining intra- or extra-axial location as well as tumour extent. Intravenous Gadolinium-DTPA (a paramagnetic contrast agent) showed tumour enhancement in the two patients in whom it was used. NMR showed more extensive abnormality than CT in nineteen patients and similar abnormality to CT in two patients. CT demonstrated calcification better than NMR in the three cases in which it was observed.

Magnetic resonance imaging of intracranial tumors in children and adolescents.

Magnetic resonance (MR) scans were reviewed of 25 children and adolescents from the age of 9 months to 18 years referred with a suspected or proven diagnosis of intracranial tumor. Twenty-one of these children had MR scans positive for tumor. Histology was available in 14. The other seven patients were managed clinically as cases of cerebral tumor, although histologic confirmation was lacking. Seventeen tumors displayed an increase in both T1 and T2. One dermoid tumor and part of another displayed a very short T1 (less than that of white matter). Two hamartomas had T1s similar to that of gray matter and a small increase in T2. Four of the children did not show MR or computed tomographic (CT) evidence of intracranial tumors. Follow-up of these cases for 1-23 months after the MR and CT studies revealed no subsequent clinical evidence of tumor. MR scans showed more extensive abnormality than did third-generation CT scans in eight of 10 cases and more extensive abnormality than EMI CT 1010 scans in 10 of 11 cases. Mass effects were better demonstrated in 14 of the 16 patients in whom they were seen. CT demonstrated calcification better than did MR in all four cases in which it was identified. The tumor-edema interface was shown better on CT in each of the three cases with contrast enhancement on CT. MR is a sensitive method of evaluating intracranial tumors in children and adolescents.