Israel Medical Association Journal

Background: Lung percutaneous needle biopsy (PNB) is routinely used to diagnose lung cancer. The most prevalent complications of PNB are pneumothorax and bleeding. Differences in characteristics of medical procedures between rural and urban hospitals are well known.

Objectives: To compare characteristics of patients and lesions between two hospitals and to evaluate whether lung PNB complications differ in rural vs. urban settings.

Methods: The authors examined case records of 561 patients who underwent lung biopsy at two different medical centers in Israel: Tel Aviv Sourasky Medical Center (urban) and Barzilai Medical Center (rural). To evaluate the complication rates, the authors analyzed findings from chest X-ray performed 2 hours after biopsy and computed tomography (CT) images at the site of biopsy.

Results: The study comprised 180 patients who underwent lung biopsy at Barzilai and 454 at Sourasky. The rate of pneumothorax did not differ between centers (12% at Barzilai and 19% at Sourasky, P = 0.08). Distance from pleura was positively correlated to pneumothorax occurrence in both centers; however, neither lesion size nor lesion locus was found to be a risk factor for pneumothorax. Mild bleeding at the biopsy site occurred equally at Barzilai and Sourasky (32% vs. 36%, P = 0.3, respectively). Furthermore, immediate CT post-biopsy at Barzilai showed 95% negative predictive value, showing that a CT scan performed immediately after lung biopsy cannot replace the routine follow-up chest X-ray in predicting iatrogenic pneumothorax.

Conclusions: CT-guided percutaneous lung biopsies are comparable between rural and urban hospitals regarding procedure characteristics and complication rates.

Background: Acute pulmonary embolism (PE) is considered to be one of the most common cardiovascular diseases with considerable mortality. Conflicting data imply possible role for echocardiography in assessing this disease.

Objectives: To determine which of the echo parameters best predicts short-term and long-term mortality in patients with PE.

Methods: We prospectively enrolled 235 patients who underwent computed tomography of pulmonary arteries (CTPA) and transthoracic Echocardiography (TTE) within < 24 hours. TTE included a prospectively designed detailed evaluation of the right heart including right ventricular (RV) myocardial performance index (RIMP), RV end diastolic and end systolic area, RV fractional area change, acceleration time (AT) of pulmonary flow and visual estimation. Interpretation and performance of TTE were blinded to the CTPA results.

Results: Although multiple TTE parameters were associated with PE, all had low discriminative capacity (AUC < 0.7). Parameters associated with 30-day mortality in univariate analysis were acceleration time (AT) < 81 msec (P = 0.04), stroke volume < 44 cc (P = 0.005), and RIMP > 0.42 (P = 0.05). The only RV independent echo parameter associated with poor long-term prognosis (adjusted for significant clinical, and routine echo associates of mortality) was RIMP (hazard ratio 3.0, P = 0.04). The only independent RV echo parameters associated with mortality in PE patients were RIMP (P = 0.05) and AT (P = 0.05). Addition of RIMP to nested models eliminated the significance of all other parameters assessing RV function.

Conclusions: Doppler-based parameters like pulmonary flow AT, RIMP, and stroke volume, have additive value in addition to visual RV estimation to assess prognosis in patients with PE.

Background: Whole-body integrated positron emission tomography / contrast-enhanced computed tomography (PET/CT) scan is increasingly used in cutaneous lymphomas. However, the value of PET/CT in the detection of cutaneous lesions in primary cutaneous B-cell lymphoma (PCBCL) has barely been investigated.

Objectives: To investigate the diagnostic accuracy of PET/CT in tracking cutaneous involvement in PCBCL.

Methods: A retrospective study was conducted on 35 consecutive patients diagnosed with cutaneous B-cell lymphoma according to the World Health Organization classification who were evaluated with PET/CT as the initial staging procedure before treatment.

Results: Thirty-five patients met the study criteria. In two patients extracutaneous disease was detected by PET/CT and CT and confirmed by biopsy. Of the 33 patients with PCBCL, 26 (79%) had small cell PCBCL (18 marginal-zone, 8 follicle-center lymphoma) and 7 (21%) had large cell PCBCL (3 follicle-center, 3 leg-type, 1 indeterminate). PET/CT detected skin lesions in 3 of 26 patients (12%) with small-cell PCBCL as compared to 6 of 7 patients with large-cell PCBLC (86%), a 7.4-fold detection risk (95% confidence interval, 2.4–22, P = 0.004). The PET-positive subgroup was characterized by larger lesion size (P < 0.001) and a higher Ki-67 proliferation index (P < 0.001).

Conclusions: The sensitivity of PET/CT for detecting cutaneous involvement of lymphomas is low for small-cell PCBCL but high for large-cell types, and thus may facilitate therapeutic strategies.

Background: Implant-related spinal infections are a surgical complication associated with high morbidity. Due to infection, hardware removal may be necessary, which could lead to pseudarthrosis and the loss of stability and alignment.

Objectives: To evaluate the accuracy and diagnostic value of ¹⁸F-fluorodeoxyglucose positron-emission tomography/computed tomography (¹⁸F-FDG PET/CT) in the workup of patients with suspected implant-related infections of the spine and to assess the clinical impact of PET/CT results on the management of these infections.

Methods: The study included nine consecutive patients with a history of spinal surgery who underwent PET/CT for evaluation of suspected spinal implant related infection. All imaging studies were performed between January 2011 and December 2013. All ¹⁸F-FDG PET/CT scans were performed on an 8 slice PET/CT following an ¹⁸F-FDG injection. Images were scored both visually and semi-quantitatively by a radiology expert. Results were compared to additional imaging studies when available, which were correlated to clinical and bacteriological findings allowing calculation of sensitivity, specificity and accuracy.

Results: Among the patients, five experienced hardware-related spinal infection. ¹⁸F-FDG PET/CT sensitivity was 80%, specificity 100%, and accuracy 88.9%. One scan produced a false negative; however, a second PET/CT scan revealed an infection.

Conclusions: PET/CT was found to be valuable for the diagnosis of postoperative hardware-related spinal infection, especially when other imaging modalities were uninformative or inconclusive. As such, PET/CT could be useful for management of infection treatment.

Background: Pulmonary embolism (PE) is the third most frequently occurring cardiovascular disease. However, the clinical presentation in patients with PE is variable.

Objectives: To evaluate the prevalence of radiological findings detected in contrast-enhanced computed tomography angiography (CTA) and their significance in patients with PE; and to assess whether the CTA findings differed in patients receiving tissue plasminogen activator (tPA) therapy from those who did not.

Methods: We retrospectively reviewed CTA scans of 186 patients diagnosed with acute PE. Incidental findings on CTA scan were assessed, including mediastinal and parenchymal lymph nodes, pleural effusion, space-occupying lesions, consolidations, emphysema, and pericardial effusion.

Results: Patients receiving tPA (19.9%) were less likely to have pleural effusion (29.7% vs. 50.3%, P = 0.024). Other CTA findings did not differ between the tPA and non-tPA groups, including lung infiltrates (40.5% vs. 38.9, P = 0.857), space-occupying lesions (5.4% vs. 6.7%, P = 1), pericardial effusion (8.1% vs. 8.7%, P = 1), emphysema (21.6% vs. 17.4%, P = 0.557), lung (18.9% vs. 24.2%, P = 0.498), and mediastinal ( 24.3% vs. 25.5%, P = 0.883) lymph nodes, respectively.

Conclusion: The prevalence of pleural effusion (unilateral or bilateral) was higher in patients not treated with tPA. Therefore, in patients with a borderline condition, the presence of pleural effusion could support the decision not to give tPA treatment.

Background: Acute non-traumatic abdominal pain is typically evaluated by abdomino-pelvic computed tomography (CT) with oral and venous contrast. The accuracy of unenhanced CT for diagnosis in this setting has not been widely studied.

Objectives: To assess the accuracy of unenhanced CT in establishing the etiology of acute non-traumatic abdominal pain.

Methods: We retrospectively reviewed the medical and imaging records of patients aged ≥ 18 years who presented to the emergency department (ED) during a 6-month period with acute non-traumatic abdominal pain of unknown etiology, and who were evaluated with non-contrast CT within 24 hours of ED admission. Clinical details were recorded. A presumptive clinical diagnosis and CT diagnosis were compared to the discharge diagnosis which was considered the reference standard. The requirement for informed consent was waived.

Results: Altogether, 315 patients met the inclusion criteria – 138 males (44%) and 177 females (56%); their mean age was 45 years (range 18–90). Clinical diagnosis correlated with the CT findings in 162 of the cases (51%). CT was accurate in 296/315 cases (94%). The leading diagnosis in cases of a mismatch between CT diagnosis and discharge diagnosis was infection mostly in the urinary tract (12/18). Sensitivity, specificity, positive predictive value and negative predictive value were 91%, 99%, 91% and 85% respectively. The discharge diagnosis was unchanged in the patients who returned to the ED within 1 week of the first admission.

Conclusions: In this study, unenhanced CT proved to be a feasible, convenient and legitimate examination for the evaluation of patients with acute non-traumatic abdominal pain presenting to the ED.

Because fragility fractures have an enormous impact on the practice of medicine and global health systems, effective screening is imperative. Currently, dual-energy X-ray absorptiometry (DXA), which has limited ability to predict fractures, is being used. We evaluated the current literature for a method that may constitute a better screening method to predict fragility fractures. A systematic review of the literature was conducted on computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound to evaluate screening methods to predict fragility fractures. We found that ultrasound had sufficient data on fracture prediction to perform meta-analysis; therefore, we analyzed prospective ultrasound cohort studies. Six study populations, consisting of 29,299 individuals (87,296 person-years of observation) and including 992 fractures, were analyzed. MRI was found to be sensitive and specific for osteoporosis, but its use for screening has not been sufficiently evaluated and more research is needed on cost, accessibility, technical challenges, and sensitivity and specificity. CT could predict fracture occurrence; however, it may be problematic for screening due to cost, exposure to radiation, and availability. Ultrasound was found to predict fracture occurrence with an increased risk of 1.45 (95% confidence interval 1.21–1.73) to fracture. Ultrasound has not replaced DXA as a screening tool for osteoporosis, perhaps due to operator-dependency and difficulty in standardization of testing.

Background: Evidence has been emerging that Helicobacter pylori may also impact colorectal cancer (CRC). Positron emission tomography/computed tomography (PET/CT) imaging can predict overall survival in CRC patients.

Objectives: To determine a possible association between H. pylori seropositivity and all-cause mortality among CRC patients evaluated by PET/CT scans.

Methods: This prospective cohort study was comprised of 110 consecutive CRC patients who had undergone a PET/CT evaluation in a tertiary academic medical center. Data included demographics, body mass index (BMI), tumor node metastasis stage at diagnosis, treatment, time from diagnosis to PET/CT, and PET/CT findings. All patients were tested for anti-H. pylori immunoglobulin G (IgG) antibodies and followed for 36 months from the day of the PET/CT scan. Mortality was documented. Univariate and multivariate Cox regression was used to estimate the hazard ratio (HR) of H. pylori serological status.

Results: During the follow-up period, of the 110 CRC patients 41 (37.3%) died and 69 (62.7%) survived. Of the 41 patients, 26 (63.4%) were H. pylori seropositive and 15 (36.6%) were seronegative. Multivariate analysis showed that H. pylori seropositivity was associated with increased mortality (HR 3.46, 95% confidence interval 1.63–7.32), stage IV at diagnosis, metastatic disease found on PET/CT, longer time from diagnosis to PET/CT, lower BMI, and older age.

Conclusions: Our findings suggest that H. pylori infection may be a risk factor for all-cause mortality among CRC patients who are evaluated by PET/CT. Multicenter studies with larger patient groups are needed to confirm our findings.