Comparison of technical success and safety of transbronchial versus percutaneous CT-guided fiducial placement for SBRT of lung tumors.

OBJECTIVE: To evaluate the technical success and safety of transbronchial (bronchoscopic) fiducial placement compared to percutaneous CT-guided fiducial placement for stereotactic body radiotherapy (SBRT) of lung tumors. MATERIALS AND METHODS: This IRB-approved, HIPAA-compliant retrospective study was performed at a single tertiary institution. Consecutive patients undergoing lung fiducial placement for purposes of guiding SBRT (CyberKnife®, Accuray, Inc.) between September 2005 to January 2013 were included in the study. Fiducial seeds were placed percutaneously with CT guidance or transbronchially with bronchoscopic guidance. We compared procedure-related complications (pneumothorax, chest tube placement), technical success (defined as implantation enabling adequate treatment planning with CT simulation) and migration rate. The need for repeat procedures and their mode was noted. Statistical analysis was performed using Fisher exact and Chi square probability tests. RESULTS: Two hundred and forty-four patients with lung tumors and 272 fiducial seed placements were included in the study. Two hundred and twenty-one of the 272 (81.2%) fiducial markers were placed percutaneously and 51/272 (18.8%) were placed transbronchially. Pneumothorax was seen in 73/221 (33%) of percutaneously-placed fiducials and in 4/51 (7.8%) of transbronchial placements (p<0.001). No significant difference was seen in the rate of chest tube placement between the two groups: 20/221 (9%) of percutaneously placed fiducials and 2/51 (3.9%) of transbronchially placed fiducials (p=0.39). Fifteen of the 51 (29%) of fiducial placements with transbronchial approach were unsuccessful, as discovered at radiotherapy planning session, and required a repeat procedure. Nine of the 15 (60%) of repeat procedures were performed percutaneously, 5/15 (33%) were placed during repeat bronchoscopy, and 1/15 (7%) was placed at transesophageal endoscopic ultrasound. No repeat fiducial placements were required for patients who had the fiducials placed percutaneously (p<0.001), with a technical success rate of 100%. CONCLUSION: Transbronchial fiducial marker placement has a significantly higher rate of failed seed placements requiring repeat procedures in comparison to percutaneous placement. Complication rate of pneumothorax requiring chest drain placement is similar between the two approaches.

Percutaneous treatment of a pancreatic fistula after pancreaticoduodenectomy.

Breakdown of the pancreaticojejunal anastomosis after a Whipple procedure is reported to occur in as many as 15% of cases. Intraoperative placement of a drain adjacent to the anastomosis is performed to allow the creation of a controlled pancreaticocutaneous fistula in the event of an anastomotic disruption. The authors present a case of successful percutaneous treatment of a disrupted pancreaticojejunal anastomosis. This was achieved with use of the resulting pancreaticocutaneous fistula for access to restore internal drainage, followed by fistula occlusion with use of gelatin pledgets.

Treatment of iatrogenic femoral pseudoaneurysms with percutaneous thrombin injection: experience in 54 patients.

PURPOSE: To assess the clinical success of ultrasonography (US)-guided thrombin injection for the treatment of iatrogenic femoral pseudoaneurysms and to identify criteria that may predispose to treatment failure. MATERIALS AND METHODS: Fifty-four iatrogenic femoral pseudoaneurysms were treated with US-guided thrombin injection. Forty-five were classified as simple (single lobe) and nine, as complex (at least two lobes and a single neck to the native vessel). Pseudoaneurysm volume, classification, thrombin dose, anticoagulation therapy status, and sheath size were compared between failed and successful cases. Seven- to 10-day follow-up US and a minimum 4-month clinical follow-up were also performed to evaluate success. RESULTS: Fifty of 54 pseudoaneurysms were successfully treated with topical thrombin without complication and included all 45 simple and five of nine complex pseudoaneurysms. US follow-up in all 50 successful cases and clinical follow-up in 37 of these revealed no recurrence. Only a complex pseudoaneurysm classification was significantly associated with failure (P<.01). Among the complex pseudoaneurysms, successful cases involved two injections and a total thrombin dose of at least 1,500 units. In failed cases, pseudoaneurysms were treated with a single injection of 1,000 units, initially thrombosed, and recurred. CONCLUSION: Simple iatrogenic femoral pseudoaneurysms, regardless of size or concomitant anticoagulation therapy, can be treated with a single injection of up to 1,000 units of topical thrombin and require no follow-up. Complex pseudoaneurysms will likely require a second injection (total thrombin dose of at least 1,500 units) and short-term clinical and US follow-up to ensure successful treatment.

Alterations in hepatic perfusion resulting from splanchnic venous luminal compromise caused by pancreatic carcinoma.

OBJECTIVE: We determined whether alterations in hepatic enhancement exist on dual phase helical CT of the liver in patients with splanchnic venous luminal compromise resulting from pancreatic adenocarcinoma. SUBJECTS AND METHODS: We examined the extent of hepatic enhancement on dual phase helical CT in 22 patients with pancreatic adenocarcinoma. Eleven patients had splanchnic venous luminal narrowing (flattening along at least 120 degrees of the circumference) of the superior mesenteric vein with (n = 3) or without (n = 8) portal vein involvement caused by tumor. In the remaining patients, splanchnic vasculature appeared normal. An additional 16 patients without pancreatic or hepatic abnormality who underwent dual phase helical CT served as control subjects. We compared the extent of arterial phase and portal venous phase enhancement among the three groups. RESULTS: The group of patients with splanchnic venous luminal compromise had significantly higher hepatic enhancement during the arterial phase (p < 0.01) and lower enhancement during the portal venous phase (p < 0.05) compared with the other two groups of patients. No significant difference in hepatic enhancement during either phase was noted between the control subjects and the patients with normal vasculature. CONCLUSION: Because hepatic enhancement correlates with perfusion, splanchnic venous luminal compromise resulting from pancreatic adenocarcinoma likely causes decreased portal venous flow and compensatory increased hepatic arterial flow. This finding supports other evidence of a homeostatic mechanism that maintains hepatic perfusion.

Iatrogenic femoral pseudoaneurysms: thrombin injection after failed US-guided compression.

Fifteen iatrogenic femoral pseudoaneurysms failed ultrasonography (US)-guided compression treatments. Despite concomitant antiplatelet or anticoagulation treatment, the 15 pseudoaneurysms were successfully and definitively treated without complication with US-guided thrombin injection. Results in this preliminary study suggest US-guided thrombin injection is a safe, expeditious, low-cost, and comfortable definitive treatment for femoral pseudoaneurysms that has advantages over both US-guided compression and open surgical repair.

Percutaneous radiofrequency tissue ablation: optimization of pulsed-radiofrequency technique to increase coagulation necrosis.

PURPOSE: To develop a computerized algorithm for pulsed, high-current percutaneous radiofrequency (RF) ablation, which maximally increases the extent of induced coagulation necrosis. MATERIALS AND METHODS: An automated, programmable algorithm for pulsed-RF deposition was designed to permit high-current deposition by periodically reducing current for 5-30 seconds during RF application. Two strategies for pulsed-RF deposition were evaluated: (i) constant peak current (900-1,800 mA) of variable duration and (ii) variable peak current (1,200-2,000 mA) for a specified minimum duration. The extent of induced coagulation was compared to results obtained with continuous (lower current) RF application. Trials were performed in ex vivo calf liver (n = 115) and in vivo porcine liver (n = 30) and muscle (n = 18) with use of 2-4-cm tip, internally cooled electrodes. RESULTS: For 3-cm electrodes in ex vivo liver, applying pulsed-RF with constant peak current for 12 minutes produced 3.5 cm +/- 0.2 of necrosis. Greater necrosis was produced with use of the variable current strategy, in which 4.5 cm +/- 0.2 of coagulation was achieved with use of an initial current > or =1,500 mA (minimum peak-RF duration of 10 sec, with 15 sec of reduced current to 100 mA between peaks; P < .01). This variable peak current algorithm also produced 3.7 cm +/- 0.6 of necrosis in in vivo liver, and 6.5 cm +/- 0.9 in in vivo muscle. Without pulsing, a maximum of 750 mA, 1,100 mA, and 1,500 mA could be applied in ex vivo liver, in vivo liver, and in vivo muscle, respectively, which resulted in 2.9 cm +/- 0.2, 2.4 cm +/- 0.2, and 5.1 cm +/- 0.4 of coagulation (P < .05, all comparisons). CONCLUSIONS: A variable peak current algorithm for pulsed-RF deposition can increase coagulation necrosis diameter over other ablation strategies. This innovation may ultimately enable the percutaneous treatment of larger tumors.

Clinically suspected pulmonary embolism: use of bilateral lower extremity US as the initial examination--a prospective study.

PURPOSE: To determine the prevalence of deep venous thrombosis (DVT) and evaluate the use of symptoms and risk factors as selection criteria in the patient population undergoing lower extremity ultrasonography (US) as an initial examination for suspected pulmonary embolism (PE). MATERIALS AND METHODS: One hundred eighty-two consecutive patients referred for bilateral lower extremity US as the first examination for suspected PE were evaluated prospectively for predisposing factors and symptoms of DVT. Patients were placed into four groups: group 1, no symptoms or risk factors; group 2, both symptoms and risk factors; group 3, only risk factors; group 4, only symptoms. The prevalence of DVT detected at lower extremity US in each group was determined. RESULTS: There were 89 patients in group 1, 12 in group 2, 43 in group 3, and 38 in group 4, with a DVT prevalence of 0%, 25%, 14%, and 24%, respectively. There was no significant difference in DVT prevalence between groups with symptoms or risk factors but a significant difference between these groups and the group lacking both symptoms and risk factors. CONCLUSION: Lower extremity US as the initial examination in patients suspected of having PE should be used only in those patients who have symptoms or risk factors for DVT. This would substantially decrease the number of examinations performed without a decline in DVT detection.

Development of virtual CT cholangiopancreatoscopy.

Endoluminal virtual computed tomographic (CT) cholangiopancreatoscopic views of the pancreatic and common bile ducts were generated in 16 patients by using helical CT data sets and endoscopic three-dimensional reconstruction images. In all patients with ductal dilatation, virtual images were obtained and findings correlated with those at helical CT and endoscopic retrograde cholangiopancreatography. Virtual CT cholangiopancreatoscopy is a feasible technique, and initial experience shows promising results.

CT detection of layering of i.v. contrast material in the abdominal aorta.

OBJECTIVE: We sought to determine whether IV contrast material would layer dependently within the abdominal aorta if administered from a location and at a rate routinely used for CT angiography. SUBJECTS AND METHODS: In 22 patients, three equal-sized regions of interest (ROIs) were symmetrically spaced across the abdominal aortic lumen at the level of the celiac axis. These positions corresponded to a dependent, a central, and a nondependent position. Time-enhancement curves were generated for each ROI after IV injection of 10 ml of nonionic contrast material, which was administered at a rate of 3.5 ml/sec, through an antecubital vein. The magnitude of and the time to peak enhancement for each ROI were determined. The 22 peak enhancement values obtained for each of the three ROIs were compared using the Newman-Keuls test. RESULTS: A statistically significant increase in peak enhancement was identified when traversing the abdominal aortic lumen from the nondependent to the central ROI (p < .05) and from the central to the dependent ROI (p < .01). Levels of mean peak enhancement were 48.3+/-19.0, 38.9+/-18.2, and 33.0+/-19.7 for the dependent, central, and nondependent ROIs, respectively. The amount of time needed to reach peak enhancement of each of the three ROIs in an individual patient was similar. CONCLUSION: A small dose of contrast material administered IV through an antecubital vein will layer dependently in the abdominal aorta--thus creating a gradient in enhancement--despite mixing in the heart. If layering of contrast material is subsequently detected during routine abdominal CT angiography, then patient positioning may need to be adjusted to maximize enhancement of targeted vessels.

Possible causes of inconclusive results on CT-guided thoracic and abdominal core biopsies.

OBJECTIVE: We sought to determine possible technical causes of inconclusive results on CT-guided core biopsies of lesions suggestive of malignancy and to determine the frequency with which such lesions are eventually found to be malignant. MATERIALS AND METHODS: We retrospectively reviewed 116 consecutive CT-guided thoracic and abdominal core biopsies performed with a 20-gauge automatic biopsy system. Biopsy results were conclusive (n = 94) if pathology confirmed malignancy and inconclusive (n = 22) if pathology results were negative for malignancy or were nondiagnostic. Lesion volume, location, number of cores, and biopsy technique (paraxial or coaxial) were compared for the conclusive and inconclusive biopsy results. Malignancy within the group of inconclusive biopsy results was determined from a second biopsy, radiographic follow-up, or surgery. RESULTS: Regression analysis identified only the biopsy method as a significant factor affecting biopsy outcome: The paraxial method was more likely to yield a conclusive result than the coaxial method (p < .002). For the two biopsy methods, lesions had similar volumes, locations, and numbers of cores obtained. For single core biopsies, both methods were equivalent. However, if two or more cores were obtained, a conclusive result was achieved in more than 90% of biopsies with the paraxial method versus 65% for the coaxial method. On follow-up, results of 14 (64%) of 22 inconclusive biopsies were malignant, indicating an overall false-negative rate of 12%. CONCLUSION: CT-guided core biopsy performed with 20-gauge automatic biopsy systems and the paraxial method will yield conclusive results significantly more often than the coaxial method. In the event of inconclusive results, malignancy will exist often enough to warrant follow-up.

Excretory phase CT urography for opacification of the urinary collecting system.

OBJECTIVE: The purpose of our study was threefold: to evaluate the ability of excretory phase CT urography to opacify the urinary collecting system by comparing opacification seen on CT with the opacification seen on a series of unmatched IV urography examinations; to determine the optimal CT urography technique for ureteral filling by comparing studies of patients who were imaged supine, prone, and with abdominal compression; and to assess the possible value that reformatted planar images might add to axial excretory phase images. SUBJECTS AND METHODS: Seventy patients with hematuria were imaged in one of four ways. Twenty-five patients underwent contrast-enhanced excretory phase helical CT of the kidneys, ureters, and bladder. All patients were imaged in a supine position. Ten other patients underwent a similar CT protocol in which we used abdominal compression. Ten further patients underwent excretory phase CT while in a prone position. A final 25 patients underwent IV urography. Each patient's collecting system was arbitrarily divided into 10 parts (both right and left sides of calices; pelvis; upper, mid, and lower ureters) for scoring of images on a five-point scale for opacification by contrast material. Opacification scores for the four groups of patients were then compared. For patients who underwent CT, reformatted images of the collecting systems were generated and evaluated for their potential to add value to the conventional axial images. RESULTS: We found no significant difference in the ability of CT urography and IV urography to yield opacification of the calices, pelvis, and upper or mid ureters. Opacification of the distal ureter was less well seen on supine CT urography than on IV urography. Prone and compression CT urography resulted in better opacification of the collecting system than the supine noncompression technique. Opacification of the distal ureter was best seen with compression CT and was as good as that seen with IV urography. Reformatted CT urography was judged to be of probable or definite additional value to the axial images in 44% of cases. In each case, we saw a pathologic finding whose relationship to the kidney and collecting system was not as easy to appreciate on the axial CT scans. CONCLUSION: CT urography with abdominal compression results in reliable opacification of the collecting system that is comparable with opacification seen on IV urography. In patients with abnormalities, reformatted images were a useful adjunct to axial images. CT urography has potential as an imaging tool for the urothelium.

The use of helical CT and CT angiography to predict vascular involvement from pancreatic cancer: correlation with findings at surgery.

OBJECTIVE: The purpose of this study was to test the accuracy of helical CT with and without CT angiography to predict vascular involvement from pancreatic cancer and correlate the resultant images with findings at surgery. SUBJECTS AND METHODS: Helical CT and CT angiography was done in 84 patients who had adenocarcinoma of the pancreas to refine an integrated CT angiography protocol. Of these patients 38 underwent radical pancreatoduodenectomy. Vessel integrity of the superior mesenteric artery, superior mesenteric vein, and portal vein was assessed prospectively on both axial CT scans and CT angiograms and correlated with findings at surgery and in particular with the ability of the surgeon to dissect the tumor from these vessels along the vascular planes. RESULTS: CT angiograms were obtained in 82 (98%) of 84 studies. Of the 38 patients who underwent radical pancreatoduodenectomy, resection was aborted in 11 (29%) because of vascular encasement, Compared with axial CT images, CT angiography was more accurate at revealing unresectable cases. In all but one of 11 patients with vessel encasement (narrowing of the lumen) revealed on CT angiography, the lesion was unresectable. Vessel encasement was shown in only two of these 11 patients on axial CT images (of these two patients, one underwent tumor resection). The negative predictive value of a resectable tumor was 96% for CT angiography and axial helical CT compared with 70% for axial helical CT images alone (p = .021). CONCLUSION: Helical CT with CT angiography of the pancreas provides useful information about local vascular involvement from pancreatic carcinoma. Compared with conventional axial helical CT, the addition of CT angiography improves radiologists' ability to predict the resectability of pancreatic tumors.

Delayed intravenous contrast medium washout from the small bowel in patients with pancreatic carcinoma and splanchnic venous invasion.

PURPOSE: Our goal was to determine whether splanchnic venous invasion by pancreatic carcinoma causes any detectable delay in contrast medium clearance from the small bowel, reflected in alterations in small bowel (SBe), portal vein (PVe), or hepatic (He) enhancement on contrast-enhanced spiral CT. METHOD: The values of SBe, PVe, and He were determined from axial images obtained during performance of an abdominal spiral CT in 20 patients with pancreatic adenocarcinoma. SBe/PVe and SBe/He, felt to reflect intravenous contrast agent washout from the small bowel wall, were also calculated. Nine patients had splanchnic venous invasion (Group 1), and 11 were free of splanchnic venous involvement (Group 2) as determined by surgery or angiography. Similar measurements and ratios were performed in 10 patients with normal spiral CT scans to serve as controls (Group 3). RESULTS: A significant increase in SBe (p < 0.05), SBe/PVe (p < 0.01), and SBe/He (p < 0.05) and decrease in PVe (p < 0.01) were found in patients with invasion when compared with those without invasion and controls, while these parameters were similar in the last two groups. Individual SBe/PVe ratios in Group 1 were consistently elevated when compared with even the highest individual values in Groups 2 and 3. CONCLUSION: A detectable increase in SBe and decrease in PVe exist when comparing patients with and without splanchnic venous invasion by pancreatic adenocarcinoma. This is possibly the result of splanchnic congestion causing a delay in contrast medium washout from the small bowel wall. The SBe/PVe ratio, felt to be an indicator of contrast agent washout, may be used as an aid during spiral CT for pancreatic adenocarcinoma to determine the status of the splanchnic vasculature in equivocal cases.

Comparison of tailored and empiric scan delays for CT angiography of the abdomen.

OBJECTIVE: Scan delays for CT angiography of the abdomen are currently individualized by matching a scan delay to the transit time of a test bolus of IV contrast material to the abdominal aorta. We sought to determine whether this procedure improves the extent and uniformity of aortic enhancement when compared with the use of a standard scan delay that is determined empirically. SUBJECTS AND METHODS: CT angiography of the upper abdomen (either the entire abdominal aorta or the aorta spanning the length of both kidneys for dedicated renal studies) was obtained in 48 patients who were considered to have normal cardiac function as determined by their referring physicians. Patients were randomized to undergo CT angiography that used either a standard scan delay of 20 or 22 sec (group 1, n = 26) or a tailored scan delay (group 2, n = 22), all other imaging parameters being identical. Tailored delays were determined by measuring contrast transit time to the abdominal aorta with software from the SmartPrep system. For each patient, aortic attenuation was measured at three different anatomic levels, and a lower threshold of 160 H was considered optimal. The number of attenuation values above this threshold, the mean attenuation, and the attenuation changes for each unit length along the abdominal aorta were then compared for the two groups of patients. RESULTS: Fifty-eight (74%) of 78 aortic attenuation values in group 1 and 55 (88%) of 69 in group 2 were above 160 H. Mean attenuation values were 192.2 H +/- 8.1 (+/-SE) in group 1 and 199.2 H +/- 10.1 in group 2. These values and the attenuation changes for each unit length along the abdominal aorta-3.7 H/cm +/- 1.7 for group 1 and 2.2 H/cm +/- 2.0 for group 2- were not significantly different. CONCLUSION: Tailoring the scan delay for CT angiography of the abdomen in patients considered to have normal cardiac function requires additional IV contrast agent, time, and patient imaging and does not improve the extent or uniformity of aortic enhancement over that obtained with a standard, empiric scan delay.

Sequential helical CT angiography of aortoiliac disease.

OBJECTIVE: The purpose of this work was to study aortoiliac disease with sequential helical CT angiography. SUBJECTS AND METHODS: Sequential helical CT angiography combines two successive helical sets for data acquisition obtained during two successive bolus injections of IV contrast material and two breath-holds. Twenty-eight patients with aneurysm and 11 with occlusive disease had CT angiography. Of those 39 patients, 18 also had conventional catheter angiography. For each of the 39 patients, a CT angiogram of three segments of the aorta and 13 arteries was assessed, including the suprarenal, juxtarenal, and infrarenal aorta; celiac axis; superior and inferior mesenteric arteries; and pairs of renal, common iliac, hypogastric, external iliac, and common femoral arteries. In 18 patients undergoing both CT and conventional angiography, the appearance of these vessels was graded as occlusive (grade 0), severely stenotic (grade 1), moderately stenotic (grade 2), mildly stenotic (grade 3), normal (grade 4), ectatic (grade 5), and aneurysmal (grade 6). RESULTS: Of the 624 arteries expected to be opacified in 39 patients, 585 (94%) were actually imaged with CT angiography. In the 18 patients who had both CT angiography and catheter angiography, the two studies were in complete agreement in 243 (90%) of 269 arteries. In 13 vessels (5%), CT angiography produced an image that was one grade higher-and in 11 vessels (4%), one grade lower-than conventional angiography. In two vessels, a two-grade difference was noted. The independent readings matched on the 0-6 scale in 95% of the evaluations. An additional 5% of the readings differed by one unit. Compared with conventional angiography, CT angiography of clinically significant (> or = 85%) narrowing (grades 0 and 1) and aneurysm (grade 6) yielded sensitivity of 93%, specificity of 96%, and accuracy of 95%. CONCLUSION: Sequential helical CT angiography of the abdomen can provide sufficient vascular detail to allow evaluation of expanded vascular territories. The technique can allow accurate assessment of both stenotic and aneurysmal disease of the aorta and the iliac arteries.