Multi--detector row helical CT of the pancreas: effect of contrast-enhanced multiphasic imaging on enhancement of the pancreas, peripancreatic vasculature, and pancreatic adenocarcinoma

PURPOSE

To determine the optimal phase for enhancement of the normal pancreas and peripancreatic vasculature and the maximal tumor-to-pancreatic parenchymal enhancement difference by using multiphase, contrast material-enhanced, multi-detector row helical computed tomography (CT).

MATERIALS AND METHODS

Forty-nine patients with a normal-appearing pancreas but suspected of having pancreatic abnormality and 28 patients with proved pancreatic adenocarcinoma underwent multiphase, contrast-enhanced, multi-detector row CT during the arterial phase (AP), pancreatic parenchymal phase (PPP), and portal venous phase (PVP). Attenuation values of the normal pancreas, pancreatic adenocarcinoma, celiac and superior mesenteric arteries, and superior mesenteric and portal veins were measured during all three imaging phases. Quantitative analysis of these measurements and subjective qualitative analysis of tumor conspicuity were performed.

RESULTS

Maximal enhancement of the normal pancreatic parenchyma occurred during the PPP. Maximal tumor-to-parenchyma attenuation differences during the PPP and PVP were equivalent but greater than that during the AP. Subjective analysis revealed that tumor conspicuity during the PPP and PVP was equivalent but superior to that during the AP. Maximal arterial enhancement was seen during the PPP, and maximal venous enhancement was seen during the PVP.

CONCLUSION

A combination of PPP and PVP imaging is sufficient for detection of pancreatic adenocarcinoma, because it provides maximal pancreatic parenchymal and peripancreatic vascular enhancement. AP imaging can be reserved for patients in whom CT angiography is required.

Hemodialysis-associated subclavian vein stenosis

This study was undertaken to evaluate hemodialysis-associated subclavian vein stenosis (SVS) and to clarify treatment of this condition. Forty-seven patients underwent upper arm venography to evaluate fistula dysfunction. Subclavian vein stenosis was documented in 12. Eleven of 12 had elevated venous dialysis pressure (196 +/- 8.9 mm Hg), and six had arm edema. All 12 had previously undergone subclavian cannulation on the side of the fistula. Thirty-five patients showed no evidence of subclavian vein stenosis. Twelve of these 35 patients (mean venous dialysis pressure 113 +/- 2.3 mm Hg) had undergone previous subclavian cannulation on the side of the fistula. The mean age of the fistula at the time of venogram in patients with subclavian vein stenosis was 17.0 months versus 5.8 months in patients with ipsilateral subclavian cannulation without subclavian vein stenosis. Percutaneous transluminal angioplasty (PTA) was performed on 11 of 12 patients with SVS lowering venous dialysis pressure and restoring patency to the fistula in 100%. Lesions recurred in two of 11 patients and were successfully retreated with PTA. We conclude that SVS is a common dialysis problem that is amenable to treatment with PTA. Elevated venous dialysis pressures are a sensitive indicator of this condition.

Delayed enhanced CT of lipid-poor adrenal adenomas

OBJECTIVE

Although representing a minority of adrenal adenomas, the lipid-poor variety cannot be accurately identified on unenhanced CT or chemical shift MR imaging. We compared the delayed contrast-enhanced CT features of lipid-poor adenomas with those of lipid-rich adenomas and of adrenal nonadenomas to determine whether there were differences in the washout features between these groups of lesions.

SUBJECTS AND METHODS

Eighteen proven lipid-poor adenomas, 56 lipid-rich adenomas, and 40 adrenal nonadenomas underwent CT before, immediately after, and 15 min delay after IV contrast injection. Region-of-interest measurements were made of all adrenal lesions at the three time points. The degree of enhancement, enhancement washout, percentage enhancement washout, and relative percentage enhancement washout were calculated for each adrenal mass. Pooled data were analyzed statistically. Optimal threshold values for diagnosing adrenal adenomas were also determined.

RESULTS

The mean CT attenuation of lipid-poor adenomas was significantly higher than that of lipid-rich adenomas at all three phases but not significantly different from that of nonadenomas. The mean percentage enhancement washout on images obtained 15 min after administration of contrast material was similar for lipid-rich and lipid-poor adenomas but was significantly higher than that of nonadenomas. The mean relative percentage enhancement washout was significantly different among all three groups.

CONCLUSION

Lipid-poor adenomas cannot be differentiated from adrenal nonadenomas on the basis of a single mean attenuation value. However, lipid-poor adrenal adenomas show enhancement and enhancement washout features nearly identical to lipid-rich adenomas and can be distinguished from nonadenomas on the basis of a percentage washout threshold value of 60% and a relative percentage washout of 40%.

Adverse reactions to gadolinium contrast media: a review of 36 cases

OBJECTIVE

We investigated the frequency, manifestations, and severity of reactions to gadolinium contrast media in patients who underwent MR imaging at a single institution.

MATERIALS AND METHODS

We reviewed the quality assurance records of 21,000 patients to whom gadolinium contrast media were administered at our institution between January 1990 and October 1994. Of these patients, 36 had adverse reactions. All 36 patients were evaluated by a radiologist at the time of the adverse reaction, and a written report of the findings was completed immediately.

RESULTS

We classified adverse reactions into four groups: mild nonallergic reactions (15 patients with nausea or vomiting), mild reactions resembling allergy (12 patients with hives, diffuse erythema, or skin irritation), moderate reactions resembling allergy (seven patients with respiratory symptoms), and life-threatening reactions resembling allergy (two patients with severe chest tightness, respiratory distress, and periorbital edema). Eleven of these 36 patients who had adverse reactions also had an unusual local reaction of skin discomfort in the extremity through which gadolinium was injected. Four of the 36 patients with adverse reactions had histories of adverse reactions to iodinated contrast media. All four of these patients and one of the remaining 32 patients with adverse reactions required treatment. No patients with adverse reactions died.

CONCLUSION

Although gadolinium contrast media are safe, we found that patients had adverse reactions at a frequency greater than we expected. Severe anaphylactoid reactions occurred in two patients (0.01%). This rate exceeds the rate of 0.0003% reported in the literature. The indexes of suspicion for the occurrence of reactions to gadolinium, and both the documentation and the management of adverse reactions, must be as rigorous for reactions associated with MR imaging contrast agents as they are for reactions associated with iodinated contrast media.

Renal adenocarcinoma: CT staging of 100 tumors

The preoperative CT scans of 100 pathologically proven renal adenocarcinomas were retrospectively reviewed to assess the accuracy of CT for staging the tumor. Regardless of tumor stage, perinephric extension was assessed with a sensitivity of 46% and a specificity of 98%. The sensitivity of CT in detecting venous invasion (either venous enlargement or thrombus) was 78%, with a specificity of 96%. For detection of metastatic adenopathy, CT had a sensitivity of 83% and specificity of 88%. Adjacent organ invasion was correctly identified in 60% of patients, with a specificity of 100%. Overall, CT correctly staged 91% of patients. If errors associated with perinephric invasion were excluded, CT staging accuracy improved to 96%. CT is useful in staging renal adenocarcinoma. If the renal vein is not well seen, angiography or sonography may be necessary to determine the presence of venous tumor extension.

Renal masses: assessment of corticomedullary-phase and nephrographic-phase CT scans

PURPOSE

To evaluate the role of thin-section helical computed tomography (CT) performed during the corticomedullary phase (CMP) and nephrographic phase (NP) of contrast enhancement in the detection and characterization of renal masses.

MATERIALS AND METHODS

Renal CT scans and medical records of 33 patients were retrospectively reviewed. In all examinations, 5-mm-thick, contiguous, helical-mode scans were obtained before and 40 seconds after initiation of dynamic bolus injection of contrast material (CMP images); 5-mm-thick, contiguous, axial-mode scans were obtained after completion of CMP scanning (NP images).

RESULTS

At review of CMP, NP, and combination images, 259, 389, and 417 lesions, respectively, were identified. The greatest difference in detection occurred in the renal medulla, with 25 lesions identified on CMP images and 111 lesions identified on NP images. False-positive results occurred when CMP images were reviewed without NP images.

CONCLUSION

CT scans obtained only during the CMP of contrast enhancement fail to depict many renal masses that are easily seen on NP images.

Different phases of renal enhancement: role in detecting and characterizing renal masses during helical CT

Although helical CT offers greater flexibility in data acquisition and reconstruction than does conventional axial CT, new opportunities for error must be understood so that optimal protocols are used that minimize the likelihood of misdiagnosis. Most nonrenal abdominal helical CT scans are obtained late in the corticomedullary phase of renal enhancement because of the preferable enhancement of other parenchymal organs. CT evaluation of the kidneys during the corticomedullary phase or at an intermediate phase between the corticomedullary and nephrographic phases has significant limitations. Therefore, dedicated renal CT performed for the detection of suspected renal masses or for the characterization of known renal masses must include images obtained during later phases of enhancement (i.e., nephrographic or excretory phase). Nephrographic or excretory phase images appear to be similar to one another but superior to corticomedullary phase images in the ability to both detect and characterize renal masses. Corticomedullary phase images should always be obtained when information about the renal vasculature is desired or when there is a possibility that a detected renal mass may represent an aneurysm or an arteriovenous malformation or fistula. Optimal technique of helical CT for staging renal cancers may require use of both corticomedullary and nephrographic or excretory phase images, although work in this area is still preliminary.

Abdominal aortic aneurysm morphology: CT features in patients with ruptured and nonruptured aneurysms

OBJECTIVE

We evaluated a variety of internal architectural features in ruptured and nonruptured abdominal aortic aneurysms to determine whether any features are associated more frequently with ruptured abdominal aortic aneurysms. These features may be useful in identifying subtle ruptures when no obvious retroperitoneal hematoma is present and may be helpful in predicting unstable aneurysms at risk for rupture.

MATERIALS AND METHODS

The CT scans of 52 patients with ruptured abdominal aortic aneurysms were reviewed and compared with those of 56 patients with asymptomatic nonruptured aneurysms exceeding 4.5 cm in diameter. All aneurysms were evaluated for size, rim calcification, thrombus amount, thrombus calcification, and lumen irregularity. In addition, four different thrombus patterns were identified and evaluated, including homogeneous, diffusely heterogeneous, periluminal halo, and crescent patterns. Statistical comparisons were adjusted for differences in size between the two groups.

RESULTS

Age, gender, and aneurysm length were not statistically different between the two groups. A larger diameter was found in the ruptured aneurysm group: 7.4 (anteroposterior) x 7.9 (transverse) cm versus 5.9 x 6.1 cm (p = .00001). More thrombus surrounded the nonruptured aneurysms (p = .014). Thrombus calcification was seen in 25% (14/56) of the control group and in 13% (7/52) of the rupture group (p = .01). Two thrombus patterns, homogeneous and periluminal halo, were encountered with similar frequencies in both groups. The diffusely heterogeneous pattern was seen more in the control group. A crescent of increased attenuation was encountered only in patients with ruptured aneurysms, at an incidence of 21% (11/52) (p = .0005). Thick and thin wall calcifications were seen in both groups, but a focal discontinuity in circumferential calcification was seen only in association with ruptured aneurysms, at an incidence of 8% (4/52) (p = .008). There was no significant difference in the number of patients whose patent lumen was irregular.

CONCLUSION

In our series, detection of a high-attenuation crescent or focal gap of otherwise circumferential wall calcification is associated with aneurysm rupture. The homogeneous, diffusely heterogeneous, and periluminal halo patterns are not specifically associated with aortic rupture. There were no significant differences in the amount of wall calcification or frequency of lumenal irregularity between patients with ruptured and those with nonruptured aneurysms.

Gastric adenocarcinoma: CT versus surgical staging

Seventy-five patients with gastric carcinoma underwent preoperative staging with computed tomography (CT). In 14 patients, CT failed to demonstrate lymphadenopathy despite the presence of malignant lymph nodes at surgery. In 13 patients, CT demonstrated enlarged nodes, but no malignant involvement was found at surgery. Although spread to the pancreas was correctly predicted in three patients with lack of fat plane between tumor and pancreas, five patients lacking a fat plane had no invasion, whereas eight patients with an intact fat plane had invasion. Thirty-five patients (47%) were incorrectly staged with CT:23 (31%) were understaged and 12 (16%) were overstaged. CT does not accurately display the true extent of disease in patients with gastric carcinoma and therefore should not be used routinely for staging.

Bladder tumor detection at virtual cystoscopy

PURPOSE

To investigate the utility of computed tomographic (CT) virtual cystoscopy in the detection of bladder tumors.

MATERIALS AND METHODS

Twenty-six patients suspected or known to have bladder neoplasms underwent CT virtual and conventional cystoscopy. The bladder was insufflated with carbon dioxide through a Foley catheter. Helical CT of the bladder was then performed. The data were downloaded to a workstation for interactive intraluminal navigation. Two radiologists blinded to the results of conventional cystoscopy independently reviewed the transverse and virtual images, with consensus readings for cases with discrepant results.

RESULTS

Thirty-six (90%) of 40 bladder lesions proved at conventional cystoscopy were detected with a combination of transverse and virtual images. Four (10%) of 40 bladder lesions, all smaller than 5 mm, were undetected. Transverse and virtual images were complementary, since six polypoid lesions smaller than 5 mm depicted on the virtual images were not seen on the transverse images. In contrast, areas of wall thickening were more readily appreciated on transverse images. CT with patients in both supine and prone positions was necessary, since seven (19%) and five (14%) of 36 lesions were seen only on supine and prone images, respectively.

CONCLUSION

CT virtual cystoscopy is a promising technique for use in bladder tumor detection of lesions larger than 5 mm. Optimal evaluation requires adequate bladder distention with the patient in both supine and prone positions and interpretation of both transverse and virtual images.

Occurrence of adverse reactions to gadolinium-based contrast material and management of patients at increased risk: a survey of the American Society of Neuroradiology Fellowship Directors

RATIONALE AND OBJECTIVES

The authors attempted to determine the frequency and severity of adverse reactions to gadolinium-based magnetic resonance (MR) contrast agents and to identify strategies for management of patients at increased risk.

MATERIALS AND METHODS

American Society of Neuroradiology program directors were surveyed about adverse reactions at their institutions to gadolinium-based contrast agents, the contrast agents responsible, and the management of patients with allergy-like reactions to iodinated or gadolinium-based agents who required MR contrast agent administration.

RESULTS

Fifty-three (50.5%) surveys were received from 105 centers. Of 687,255 gadopentetate dimeglumine injections, 314 (0.046%) nonallergic reactions and 107 (0.016%) mild, 28 (0.004%) moderate, and five (0.001%) severe allergy-like reactions occurred. Of 74,275 gadodiamide injections, 11 (0.015%) nonallergic and 12 (0.016%) mild allergy-like reactions occurred. Of 64,005 gadoteridol administrations, 171 (0.267%) nonallergic reactions and 49 (0.077%) mild, 29 (0.047%) moderate, and 11 (0.017%) severe allergy-like reactions occurred. Twenty-six departments took no precautions for patients with previous allergy-like reactions to iodinated contrast material. Nineteen did not premedicate patients who previously had reactions to gadolinium-based agents before repeat administration of MR contrast agents.

CONCLUSION

Although MR contrast agents are safe, adverse reactions occur. Many centers have not adopted policies for the OFF

Computed tomography of renal lymphoma

The CT studies of 29 patients with renal or perirenal lymphoma were retrospectively reviewed. Four patterns of disease were identified. Seventeen of 29 patients (59%) had bilateral renal masses. Only seven of these patients had associated enlarged retroperitoneal lymph nodes. Eight patients (28%) had single renal or perirenal lesions adjacent to or contiguous with bulky retroperitoneal lymphadenopathy. Three patients had infiltration of the perirenal space without significant renal parenchymal involvement, and one patient had a solitary renal mass. No patients in this series had diffuse involvement of the kidney without a focal mass. Renal involvement with lymphoma should be considered in any patient who develops multiple homogeneous solid renal or perirenal masses, even in the absence of other retroperitoneal disease.

Multislice CT urography: state of the art

Recent improvements in helical CT hardware and software have provided imagers with the tools to obtain an increasingly large number of very thin axial images. As a result, a number of new applications for multislice CT have recently been developed, one of which is CT urography. The motivation for performing CT urography is the desire to create a single imaging test that can completely assess the kidneys and urinary tract for urolithiasis, renal masses and mucosal abnormalities of the renal collecting system, ureters and bladder. Although the preferred technique for performing multislice CT urography has not yet been determined and results are preliminary, early indications suggest that this examination can detect even subtle benign and malignant urothelial abnormalities and that it has the potential to completely replace excretory urography within the next several years. An important limitation of multislice CT urography is increased patient radiation exposure encountered when some of the more thorough recommended techniques are utilized.

Retroperitoneal plexiform neurofibromas: CT findings

OBJECTIVE

CT findings in retroperitoneal plexiform neurofibromas have been described previously only in a few case reports. This study was performed to characterize the CT features in a larger group of patients with this benign neoplasm and to delineate those findings that suggest the presence of concurrent malignant tumors of the nerve sheath.

MATERIALS AND METHODS

From a review of CT scans showing retroperitoneal masses in 25 patients with neurofibromatosis type 1, we identified 16 patients whose clinical and CT features suggested the diagnosis of plexiform neurofibromas. The diagnosis was confirmed by biopsy of the mass in three patients. In the remaining 13 patients, the diagnosis was supported by long-term follow-up in eight and by biopsy confirmation of either a neurofibroma or a plexiform neurofibroma elsewhere in the body in the other five patients. Concurrent retroperitoneal malignant tumors of the nerve sheath were confirmed by biopsy in five patients.

RESULTS

CT showed bilateral, remarkably symmetric masses extending along the medial and posterior aspects of the psoas muscles in 15 of the 16 patients. Parapsoas masses tended to be long and cylindrical. In addition to parapsoas masses, eight patients had bilateral, symmetric masses extending along the anterior aspect of the sacrum in a sheetlike fashion. Masses were identified in the region of the celiac axis and the origin of the superior mesenteric artery in four patients. Only one patient had an isolated celiac/superior mesenteric artery mass without associated parapsoas or presacral masses. Retroperitoneal masses were of homogeneous, low attenuation when compared with adjacent muscle on CT images of 14 patients. Two patients had an unusual swirling and serpiginous pattern of increased attenuation superimposed on a low-attenuation background. Asymmetry in size and attenuation was due to malignant nerve sheath tumors in five patients.

CONCLUSION

Retroperitoneal plexiform neurofibromas have a characteristic appearance on CT scans. They are typically bilateral, symmetric, low-attenuation masses in a parapsoas or presacral location. Asymmetry in size and attenuation of a larger mass suggests the possibility of a malignant tumor of the nerve sheath. Recognition of the CT features of a retroperitoneal plexiform neurofibroma can obviate the expense, pain, and risk of an unnecessary biopsy. Conversely, detection of findings suggestive of malignant tumor can lead to appropriate recommendation of biopsy.

CT of abdominal aortic aneurysms

CT often plays an important role in the evaluation of abdominal aortic aneurysms. Many vascular surgeons consider it an integral part of the preoperative assessment of aneurysms. CT is also used to examine symptomatic patients with possible aneurysmal ruptures when the diagnosis is unclear and to study patients with aortic grafts in whom potentially lethal complications, such as graft infection, aortoenteric fistulas, and anastomotic pseudoaneurysms, are clinically suspected. This paper reviews the current role of CT in the assessment of patients with abdominal aortic aneurysms.

Lumbar hernia: diagnosis by CT

Lumbar hernias occur in the region of the flank bounded by the 12th rib, the iliac crest, and the erector spinae and external oblique muscles. We present the CT findings of seven lumbar hernias: six traumatic (four secondary to postoperative flank incisions, one secondary to an iliac bone-graft donor site, one secondary to nonunion of an iliac fracture) and one spontaneous. Because CT portrays the anatomic relationships in this region so well, it may be the only radiographic procedure necessary to make the diagnosis of a lumbar hernia. Furthermore, it can be helpful in the assessment of symptomatic patients after flank incision, to differentiate postincisional muscular weakness and intercostal neuralgia from a lumbar hernia.

Cystic renal masses: a reevaluation of the usefulness of the Bosniak classification system

RATIONALE AND OBJECTIVES

We evaluated the utility of the Bosniak system for classifying cystic renal masses on computed tomography (CT) scans.

METHODS

The CT scans of 20 patients with 24 cystic renal masses that were subsequently surgically removed or biopsied were reviewed retrospectively. Masses were categorized using the Bosniak system and were correlated with the pathology results.

RESULTS

The final pathology results of the cystic renal masses were as follows: Seven of seven category I lesions were benign, one of five category II lesions was benign, zero of four category III lesions were benign, and zero of six category IV lesions were benign. Neither of two unclassifiable cystic lesions were benign. The average enhancement of lesions in categories II, III, and IV was 6.3, 2.3, and 27.6 Hounsfield units (H), respectively. The two uncategorizable lesions had a mean enhancement of 26.8 H.

CONCLUSION

The results of our study serve to underscore some limitations of the Bosniak classification system because most of our category II and all of our category III lesions were malignant, suggesting that minimally complex cystic renal masses may contain malignant cells. Contrast enhancement of less than 10 H was demonstrated in lesions in categories II and III.

Sensitivity of CT scout radiography and abdominal radiography for revealing ureteral calculi on helical CT: implications for radiologic follow-up

OBJECTIVE

We compared the sensitivity of CT scout radiography with that of abdominal radiography in revealing ureteral calculi on unenhanced helical CT.

MATERIALS AND METHODS

Over a 6-month period, patients presenting to the emergency department with acute flank pain were examined with standard abdominal radiography and unenhanced helical CT, which included CT scout radiography. In 60 patients in whom a diagnosis of ureteral calculus was made, CT scout radiographs and abdominal radiographs were examined by two interpreters who assessed whether stones could be visualized. All CT scout radiographs were viewed on a workstation using optimized window settings.

RESULTS

CT scout radiography and abdominal radiography revealed 28 (47%) and 36 (60%) of 60 ureteral calculi, respectively. All ureteral calculi that appeared on CT scout radiography also appeared on abdominal radiography. However, eight calculi that were visible on abdominal radiography were not visible on CT scout radiography. CT scout radiography and abdominal radiography revealed 28% and 46% of 39 calculi less than or equal to 3 mm in diameter, respectively. For 21 calculi larger than 3 mm, the sensitivity of CT scout radiography and abdominal radiography was 81% and 86%, respectively.

CONCLUSION

Abdominal radiography is more sensitive than CT scout radiography in revealing ureteral calculi; however, some calculi revealed on unenhanced helical CT cannot be seen on either abdominal radiography or CT scout radiography. Ureteral calculi not visible on either study can only be followed, when necessary, with unenhanced helical CT.

Retroperitoneal fibrosis: use of CT in distinguishing among possible causes

Retroperitoneal fibrosis (RF) often presents with characteristic clinical and urographic findings that suggest the diagnosis. However, malignant adenopathy may mimic this appearance. CT examinations of 10 patients with medial deviation of the ureters suggesting retroperitoneal fibrosis were reviewed. Six had idiopathic RF, two had perianeurysmal fibrosis, and two had malignant retroperitoneal adenopathy. The CT findings in idiopathic RF included a homogeneous mass enveloping the ureters, aorta, and inferior vena cava. The aortic aneurysm was easily seen in both patients with perianeurysmal fibrosis. In one of the two patients with malignant retroperitoneal adenopathy, the CT findings mimicked idiopathic RF. In the other patient, the nodes involved with metastatic melanoma caused significant anterior displacement of the aorta and inferior vena cava. This marked anterior displacement of the great vessels is not seen in idiopathic RF, and may be useful in differentiating these entities.

Extravascular toxicity of two magnetic resonance contrast agents. Preliminary experience in the rat

We compared the relative toxicities of standard concentrations of two magnetic resonance imaging (MRI) contrast agents, ionic gadolinium diethylenetriaminepentacetic acid (DTPA) and low-osmolar gadolinium-1, 4, 7 tris (carboxymethyl)-10-(2'-hydroxypropyl)-1, 4, 7, 10 tetra-azacyclododecane (HP-DO3A) with that of the conventional radiographic contrast medium meglumine diatrizoate, when extravasated into the deep dermal tissues of laboratory rats. Gadolinium-DTPA caused moderate necrosis, hemorrhage, and edema which was not statistically different than meglumine diatrizoate. In contrast, gadolinium HP-DO3A was significantly less toxic than meglumine diatrizoate. Additional experience will be needed in order to determine whether these laboratory results will be clinically relevant in humans.

Extravasation of nonionic radiologic contrast media: efficacy of conservative treatment

The authors report four large-volume (greater than 20-mL) extravasations in adults and one intermediate-volume (18-mL) extravasation in an infant of iopamidol, a nonionic contrast medium, that occurred during dynamic enhanced computed tomography. The largest volume that extravasated (in two patients) was 150 mL. Although all five patients initially had swelling, erythema, and/or pain after extravasation, all recovered uneventfully. The authors suggest that emergent surgical drainage of all large-volume extravasations, a procedure that was recently recommended, is unnecessary. Large extravascular collections of nonionic contrast media can be well tolerated and are often resorbed without adverse sequelae.

Transthoracic needle aspiration: use of a small chest tube to treat pneumothorax

The primary complication of transthoracic needle aspiration is pneumothorax. The efficacy and safety of using a small chest tube to treat this complication were examined by reviewing the records of 876 patients who underwent transthoracic needle aspirations between January 1981 and February 1986. Among these patients, 212 (24%) sustained a pneumothorax, and 92 (11%) required placement of a small 9-French chest tube attached to a flutter-type (Heimlich) valve. Duration of chest-tube drainage ranged between 24 hr and 3 weeks (mean, 2.2 days). Complete resolution of the pneumothorax and subsequent removal of the chest tube after 24 hr of drainage occurred in 38 (41%) of the 92 patients. Twenty-nine (32%) required 48 hr of drainage, and nine (10%) required 3 days. The remaining 16 (17%) required longer periods of drainage ranging from 4 days to 3 weeks. The tubes of six of this last group of patients were attached to a suction apparatus, and three of these patients eventually had a 28-French chest tube placed surgically. No significant complications occurred. The use of a small chest tube for treatment of pneumothorax after transthoracic needle aspiration is easy, safe, and efficacious.

Ovarian vein thrombosis in oncology patients: CT detection and clinical significance

During a 13-month period, ovarian vein thrombosis (OVT) was detected with CT in six patients with malignant tumors, five of whom were receiving high-dose chemotherapy at the time of diagnosis. None of these patients belonged to previously recognized risk groups. Unlike most patients with postpartum or postoperative OVT, only one of our patients was symptomatic, and none had related CT findings, such as uterine enlargement or other enhancing pelvic masses. Patients with malignant tumors, particularly those undergoing chemotherapy, are at risk for developing OVT. As OVT is often asymptomatic in these patients, and thrombus may resolve without treatment, anticoagulation may not be routinely necessary.