Israel Medical Association Journal

Background: Plain abdominal radiographs are still performed as a first imaging examination to evaluate abdominal pain in the emergency department (ED), despite uncertainty regarding their utility.

Objectives: To describe the frequency and outcomes of the use of plain abdominal radiographs in the diagnosis of patients presenting with acute non-traumatic abdominal pain in the ED of a medical center. 

Methods: We retrospectively reviewed the records of patients presenting to the ED with acute abdominal pain during a 6 month period. Further imaging (computed tomography, ultrasonography), when performed, was compared with the abdominal radiography. 

Results: Of 573 consecutive patients, 300 (52%) underwent abdominal radiography. Findings were normal in 88% (n=264), non-specific in 7.3% (n=22), and abnormal in 4.7% (n=14). For those with normal results, no further imaging was ordered for 43% (114/264). Of the 57% (150/264) who had follow-up imaging, 65% (98/150) showed abnormal findings. In 9 (3%) of the 300 patients, abdominal radiography identified bowel perforations and obstructions, and treatment was provided without the need for further radiologic examination.

Conclusions: The use of plain abdominal radiography is still common despite the high rate of false positive results. Efforts are needed to decrease the indiscriminate use of radiography in patients presenting with abdominal symptoms.

 

Background: The practice of administering intravenous contrast to children varies by institution depending on their routine. 

Objectives: To assess the necessity of routine contrast administration in brain magnetic resonance imaging (MRI) of pediatric outpatients referred for chronic headache workups. 

Methods: We conducted a retrospective review of consecutive pediatric brain MRI examinations performed during January and February 2014 in 30 pediatric outpatients referred for evaluation of chronic headache. Independent review was performed by two board-certified neuroradiologists. The raters reviewed each MRI first as a non-contrast examination (without seeing the post-contrast images) and then with post-contrast images. 

Results: No abnormalities were found in six patients. One patient had an indeterminate finding of a tubular cerebellar lesion requiring follow-up. In the remaining patients (n=23), the findings were subclinical and included: mucosal thickening in the paranasal sinuses in 9 patients, cystic changes of the pineal gland in 8 (size 2–9 mm), small developmental venous anomalies in 6, non-specific FLAIR hyperintensities in 4, opacification of the mastoids in 2, and telangiectasia in 1 patient. The subclinical cases that were missed on pre-contrast images were: one small developmental venous anomaly, one telangiectasia and one small pineal cyst, none of which hold clinical significance. All kappa inter-rater and intra-rater agreement scores resulted in values above 0.75, excellent agreement according to Fleiss guidelines.

Conclusions: There seems to be little reason to medically justify large-scale use of routine IV contrast administration to evaluate a brain MRI of pediatric patients referred for chronic headache. 

 

Breast magnetic resonance imaging (MRI) has an important role in the detection and diagnosis of breast cancer. Suspicious findings on MRI are further evaluated with ultrasound. This case series illustrates the use of automated breast volumetric ultrasound (ABVS) as a tool for second-look ultrasound (SLUS) following MRI. Seven women underwent breast MRI with findings necessitating SLUS. ABVS was used for second look and all MRI lesions were detected. Four cancers, one fibroadenoma and two benign lesions, were diagnosed. This case series shows that ABVS can be used as a tool for SLUS following MRI and in some cases is superior to hand-held ultrasound.

Abstract

Background: Only a few studies have compared the accuracy of magnetic resonance enterography (MRE) and computed tomography enterography (CTE) in the diagnosis of Crohn's disease and its complications.

Objectives: To compare the sensitivity of MRE and CTE analysis in their ability to detect, sign-by-sign, 10 classical imaging signs of Crohn's disease.

Methods: The study group comprised 42 biopsy-proven Crohn's disease patients who underwent both CTE and MRE within an average period of 6 weeks. Agreement between the two modalities in detecting the 10 most significant radiological signs of CD was evaluated using the Kappa index. The sensitivity of MRE and CTE was calculated using a standard of reference composed of all the findings seen by CTE and/or MRE. We analyzed MRE and CTE sensitivity separately in two groups, according to the time interval between the examinations.

Results: Agreement between CTE and MRE was more than 70% in 8 of the 10 signs: mural thickening, phlegmon, stenosis, skip lesions, mucosal stratification, fistula, abscess, and creeping fat. The Kappa level of agreement values for CTE versus MRE varied between substantial for phlegmon and skip lesions; moderate for fistula, creeping fat, abscess and mural thickening; and fair for stenosis and dilatation. CTE detected more findings than MRE, except for creeping fat and fistula. There was no significant difference in the sensitivity of CTE and MRE in the two groups defined by the time interval (time < 1.5 and time > 1.5 months) except for detection of dilatation.

Conclusions: Almost all imaging signs of Crohn's disease were detected equally well by both modalities regardless of the time interval between examinations. We therefore consider MRE to be reliable for imaging and follow-up in patients with Crohn's disease who may need recurrent imaging.

Abstract

Background: Forensic imaging was officially introduced in Israel in 2011. Religious and cultural opposition to autopsies prevails in most of the population in Israel.

Objectives: To examine the extent to which forensic imaging has been accepted as an adjuvant or partial replacement of forensic autopsy, particularly among those opposed to forensic autopsy.

Methods: The study was conducted in a pediatric population. Data were collected from the National Center of Forensic Medicine and Assaf Harofeh Medical Center during the 18 month period following the introduction of forensic imaging (group A). The data were compared to those of the previous 18 months (group B). The examined parameters were cases submitted, examined, autopsied or imaged depending on family consent.

Results: Consent to autopsy was similar in both groups (A = 56% vs. B = 54%). In group A, consent for imaging was 24% of all cases, and of those imaged 77% underwent autopsy. Of those examined externally only, 16% consented to imaging. For 7% of the total cases in group A, estimation of cause of death was based on virtopsy alone.

Conclusions: In a country with a high level of religious opposition to autopsy, it is a challenge to add forensic to the pediatric forensic investigation. Those consenting to forensic imaging are more likely to be those consenting to autopsy. Consent for forensic imaging only was given in 7% of cases. Greater efforts should be invested to educate and inform the public regarding the benefits of virtual autopsy and the importance of data acquired from forensic images. 

Background: Medical evaluation of a suspected meniscus injury begins with a history-taking and physical examination. Suspected meniscus injuries not responding to treatment are usually sent for imaging to confirm the diagnosis before arthroscopy. Tc-MDP bone single photon emission computed tomography (SPECT) scan has been suggested as an alternative to magnetic resonance imaging (MRI) in evaluating suspected knee meniscus tears.

Objectives: To examine the accuracy of knee SPECT as a tool to identify meniscus tears versus that of MRI as compared to the gold standard of arthroscopy.

Methods: The Israel Defense Forces database for 2005 through 2009 was searched according to the key words knee MRI, knee SPECT and knee arthroscopy. We identified 330 subjects who had undergone both a single knee SPECT and a single knee MRI prior to knee surgery. The medical files of 193 of the 330 subjects were randomly selected for review. A comparison was made between the preoperative SPECT and MRI studies and the arthroscopic finding. The sensitivity, specificity and accuracy were calculated.

Results: The subjects’ age was 21.3 ± 3.9. The agreement between SPECT and arthroscopy was 0.14 for the medial meniscus and 0.29 for the lateral meniscus. The agreement between MRI and arthroscopy was 0.59 for the medial meniscus and 0.69 for the lateral meniscus. SPECT scan was found to be 61% sensitive, 54% specific and 58% accurate in detecting common knee pathology, whereas MRI was found to be 95% sensitive, 67% specific and 85% accurate.

Conclusions: Knee SPECT has a lower sensitivity, specificity and accuracy than MRI in evaluating meniscal injuries and its use can result in increased unnecessary surgery.

Objectives: To investigate the association between CAD and H. pylori infection using MPI.

Methods: This prospective study evaluated CAD positivity among consecutive patients referred to a tertiary medical center for a stress/rest MPI. All patients were tested for serum anti-H. pylori and CagA protein immunoglobulin G antibodies. The CAD-positive group included patients with ischemia and/or myocardial infarctions (MI) on a stress MPI, coronary artery bypass graft surgery (CABG) or percutaneous coronary interventions (PCI). CAD-negative subjects were defined as participants with a normal MPI, no pathological Q waves in resting ECG tracing, and no history of CAD. Both groups were compared for H. pylori and CagA seropositivity. Patients’ demographic data, risk factors for CAD, and childhood socioeconomic status were recorded.

Results: The study group consisted of 300 consecutive patients, 170 men and 130 women; 64% (110/173) CAD-positive patients and 47% (60/127) CAD-negative participants were found seropositive for H. pylori infection (P = 0.005). In the adjusted analysis, H. pylori infection was found to be associated with CAD- positive (odds ratio 1.83, 95% confidence interval 1.06–3.17, P = 0.031), and MI (fixed perfusion defects on MPI) (OR 3.36, 95%CI 1.44–7.84, P = 0.005). No association was noted with CagA positivity.

Conclusions: In patients undergoing a stress MPI, serum anti-H. pylori antibodies positivity was found to be associated with CAD, independent of traditional cardiovascular risk factors. 

Objectives: To measure the tumor-to-breast volume ratio using MRI in order to assess whether there is a correlation between this ratio and the type of surgery selected (breast-conserving or mastectomy).

Methods: The volumes of the tumor and the breast and the tumor-to-breast volume ratio were retrospectively calculated using preoperative breast MRI in 76 patients who underwent breast-conserving surgery or mastectomy.

Results: Breast-conserving surgery (lumpectomy) was performed in 64 patients and mastectomy in 12. The average tumor-to-breast volume ratio was 0.06 (6%) in the lumpectomy group and 0.30 (30%) in the mastectomy group (P < 0.0001).

Conclusion: The tumor-to-breast volume ratio correlated with the type of surgery. As measured on MRI, this ratio is an accurate means of determining the type of surgery best suited for a given patient. It is recommended that MRI-determined tumor-to-breast volume ratio become part of the surgical planning protocol for patients diagnosed with breast cancer.