The authors describe a case in which a patient underwent percutaneous cryoablation of a suspected right renal cell carcinoma complicated by bleeding. Urgent angiography revealed a lower renal pole arteriovenous (AV) fistula, correlating with the recent treatment site. This AV fistula was successfully treated with coil and Gelfoam embolization. Three days later, the patient's hemoglobin dropped following dialysis. Computed tomography (CT) imaging revealed an increase in the size of the pararenal hematoma. There were multiple pseudoaneurysms as well as a small AV fistula on repeat angiography. The right main renal artery was coil embolized.
Purpose:
In follow‐up T2‐weighted MR images of spinal tumor patients treated with stereotactic body radiation therapy (SBRT), high intensity features embedded in dark surroundings may suggest a local failure (LF). We investigated image intensity histogram in imaging features to predict LF and local control (LC).
Methods:
Sixty‐seven spinal tumors were treated with SBRT at our institution with scheduled follow‐up MR T2‐weighted (TR 3200–6600ms; TE 75‐132ms) imaging. The LF group included 10 tumors with 8.7 months median follow‐up, while the LC group had 11 tumors with 24.1 months median follow‐up. The follow‐up images were fused to the planning CT. Image intensity histograms of the GTV were calculated. Voxels in greater than 90% (V90), 80% (V80), and peak (Vpeak) of the histogram were grouped into sub‐ROIs to determine the best feature histogram. The intensity of each sub‐ROI was evaluated using the mean T2‐weighted signal ratio (intensity in sub‐ROI / intensity in normal vertebrae). An ROC curve in predicting LF for each sub‐ROI was calculated to determine the best feature histogram parameter for LF prediction.
Results:
Mean T2‐weighted signal ratio in the LF group was significantly higher than that in the LC group for all sub‐ROIs (1.1±0.4 vs. 0.7±0.2, 1.2±0.4 vs. 0.8±0.2, 1.4±0.5 vs. 0.8±0.2, for V90, V80, and Vpeak, p=0.02, 0.02, and 0.002, respectively). The corresponding areas‐under‐curve (AUC) of ROC were 0.78, 0.80, and 0.87, p=0.02, 0.03, 0.004, respectively. No correlation was found between T2‐weighted signal ratio in Vpeak and follow‐up time (Pearson's ρ=0.15).
Conclusion:
Increased T2‐weighted signal can be used to identify local failure while decreased signal indicates local control after spinal SBRT. By choosing the best histogram parameter (here the Vpeak), the AUC of the ROC can be substantially improved, which implies reliable prediction of LC and LF. These results are being further studied and validated with large multi‐institutional data.
Preceding the advent of effective antimicrobial regimens, "collapse therapy" was the mainstay of treatment for tuberculosis. This therapy became obsolete after the discovery of potent multi-drug regimens in the 1950s. However, collapse therapy represents an important historical treatment that was lifesaving to many patients in the 1930s and 1940s. CASE PRESENTATION: A 97-year-old female presented to the emergency department with acute onset dyspnea, worsening over several days. She was tachycardic to 163 bpm. Physical examination revealed right hemithoracic basilar crackles and poor air movement over the entire left hemithorax. EKG revealed multi-focal atrial tachycardia. Troponins were negative, however pro-BNP was elevated at 2950 ng/L. Chest radiography revealed left hemithoracic mass-like opacification with calcific margins. Chest computed tomography confirmed a well-circumscribed, peripherally calcified left hemithoracic mass. Contents of the mass were heterogeneous with areas measuring-60 HU, suggestive of lipoid material. Right-sided pleural effusion was also noted. She disclosed that she had been diagnosed with tuberculosis in her twenties and spent 5 months in a sanatorium prior to undergoing surgery. She denied post-operative complications. Her dyspnea resolved with diuresis and rate control. She was discharged home in improved condition.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.