Early Venous Phase Research Articles

Sclerosing Angiomatoid Nodular Transformation

AJR:192, May 2009 known, and a strong correlation with another disease process has not been found [3, 4]. SANT has a distinctive gross appearance. It is a well-circumscribed mass composed of numerous nodules derived from red pulp alternating with bands of dense fibrous tissue that coalesce to form a central, stellate fibrous scar [3]. Microscopically, SANT is characterized by multiple nodular aggregates of plump endothelial cells and pericytes lining prominent slitlike vascular spaces [2–5]. The nodules are associated with extravasated erythrocytes, and the intervening collagenous stroma contains variable numbers of reactive myofibroblasts, hemosiderin-laden macrophages, lymphocytes, and plasma cells. The endothelial cells exhibit minimal cytologic atypia, with rare mitotic activity [2–5]. Limited imaging information is conveyed by the pathology literature. Two articles provide correlative enhanced CT images [4, 5]. Both have images or descriptions depicting a hypodense mass in the early portal venous phase, with areas of peripheral enhancement that become homogeneous on more delayed imaging. Description of MR findings in one case indicates that the mass showed low signal on T1 and T2 sequences [4]. A recent report, however, describes a single case showing early peripheral enhancement progressing centripetally in a radial pattern [6]. Although the small number of cases with radiologic description precludes definitive assessment, suggestive characteristics of SANT include a well-circumscribed solitary mass of the spleen that initially shows low attenuation with areas of peripheral enhancement after the administration of contrast material. This may reflect the gross pathologic appearance of central fibrous stroma and more peripheral coalescent nodules of angiomatoid Sclerosing Angiomatoid Nodular Transformation

What is a reliable CT scan for diagnosing splenosis under emergency conditions?

Splenosis is a condition in which splenic tissue is present in a non-anatomical position. Implants of splenic tissue can mimic neoplasms and only specific examinations can confirm the correct diagnosis. Here we report a case of a 23-year-old male patient with a history of surgical splenectomy during childhood after trauma. He was admitted to the emergency department with acute bowel obstruction. An abdominal-pelvic computed tomography (CT) scan revealed small bowel obstruction and the presence of two rounded, solid masses located in the rectal-vescical pouch. Quantitative analyses of the different density values in the arterial phase and early portal venous phase demonstrated that these lesions were highly vascularised (92 and 97 Hounsfield Units, respectively). The hypothesis of an ectopic splenic mass was made after evaluation of the CT images and clinical history. The acute bowel obstruction caused by adhesive intestinal syndrome was resolved by surgical adhesiolysis. The smallest mass adherent to the rectum was removed. Histopathologic examination confirmed the benign nature of the lesion, which consisted of splenic tissue.

CORRELATION OF FUNDUS AUTOFLUORESCENCE WITH FLUORESCEIN AND INDOCYANINE GREEN ANGIOGRAPHY IN CHOROIDAL MELANOCYTIC LESIONS

To correlate fundus autofluorescence (FAF) patterns with fluorescein/indocyanine green angiographic (FA/ICGA) features in choroidal melanocytic lesions. Retrospective chart review of 30 consecutive patients with choroidal nevi and melanoma who underwent FAF photography and FA/ICGA. The FAF pattern was classified as patchy or diffuse. The FA images were evaluated at the arterial, early venous, late venous, midphase, and late phases. The ICGA images were evaluated at the early and midlate phase. The fluorescence within the tumor was classified as hyperfluorescent, pinpoint hyperfluorescent, isofluorescent, or hypofluorescent with respect to the surrounding retina or choroid. Statistical analysis was performed using two sample t test for continuous data. For categorical or ordinal data, Pearson chi-square or Fisher's exact test was used depending on the sample size being studied. Nineteen of 30 tumors (63.3%) were choroidal melanoma and 11 (36.7%) were choroidal nevus. Thirteen choroidal melanomas had a diffuse FAF pattern. Six choroidal melanomas and 11 choroidal nevi had a patchy FAF pattern. The diffuse FAF pattern was significantly associated with the clinical diagnosis of choroidal melanoma versus choroidal nevus (P = 0.00001), increased tumor thickness (P = 0.00001), and increased tumor base diameter (P = 0.001), partially pigmented or amelanotic versus pigmented lesion color (P = 0.006), early venous hyperfluorescence on FA (P = 0.015), and late hyperfluorescence on FA (P = 0.018). Diffuse FAF is more often associated with larger choroidal melanomas as well as early venous and late hyperfluorescence on FA angiography.

Severe thrombocytopenia caused by littoral cell angioma

In September 2002, a 54-year-old man, who had been on hemodialysis because of chronic renal failure due to polycystic kidney disease, referred to our department because of progressing thrombocytopenia with platelet count of 27 9 10/L. Abdominal ultrasonography showed significant splenomegaly with numerous hyperechoic lesions of various sizes which might cause the thrombocytopenia. Computed tomographic (CT) scan revealed that multiple hypoenhanced lesions at the early portal venous phase became isoattenuating during the late portal venous Fig. 1 CT scans, microscopic and immunohistochemical images of littoral cell angioma

Retroaortic left renal vein: multidetector computed tomography angiography findings and its clinical importance

To evaluate the appearance, type, frequency, and clinical importance of retroaortic left renal vein (RLRV) in patients examined with multidetector computed tomography (MDCT) angiography. A total of 1856 patients who underwent CT with urological symptoms (hematuria, flank and abdominal pain, left gonadal vein varicocele) (n = 889) or with symptoms other than urological ones (n = 967) were prospectively evaluated for the presence of RLRV. CT was performed with 110 ml of iodinated contrast material through the antecubital vein at a rate of 3.5 ml/s. Late arterial and early venous phase volumetric data sets were acquired at 30 and 65 s, respectively, from the start of the intravenous injection of contrast medium. In addition to axial images, multiplanar reconstructions (MPR), maximum-intensity projection (MIP), and three-dimensional volume-rendering (3D VR) images were used to assess left renal vein anomalies. Left renal vein anomalies were classified into four types according to their appearance: I) RLRV joining the inferior vena cava (IVC) in the orthotopic position; II) RLRV joining the IVC at level L4-L5; III) circumaortic or collar left renal vein; IV) RLRV joining the left common iliac vein. RLRV was detected in 68 (3.6%) of the 1856 patients, with 26, 22, 17, and three of types I, II, III, and IV, respectively. Forty-four of the 68 patients with RLRV (65%) were in the group with urological symptoms, while 24 patients (35%) were in the group without urological symptoms. Compression of the RLRV was found in 16 patients in the urological symptoms group, while compression was detected in only three patients in the other group. This difference was statistically significant (P<0.05). The most common urological symptom was hematuria. The frequency of urological symptoms was higher in groups II and IV compared to the other groups. MDCT angiography with axial, MPR, MIP, and 3D VR images is effective in the detection of vascular renal anomalies such as RLRV. Diagnosing RLRV and differentiating it from other pathologic conditions causing hematuria is important in order to avoid complications during retroperitoneal surgery or interventional procedures.

The Fat Halo Sign

The inner layer of soft-tissue attenuation represents the bowel (small and/or large) mucosa, while the layer of low attenuation ( 18 to 64 HU) results from widening and fatty infiltration of the submucosa. The outer soft-tissue attenuation layer represents the muscularis propria and serosa (2,4,6–8). The fat halo sign can be depicted on CT scans obtained without intravenous (IV) contrast material because of the marked differences in tissue attenuation (3). However, the different layers of attenuation can also be appreciated during the late arterial and early portal venous phases of IV contrast enhancement (3,7).

Serial Contrast-Enhanced MRI of the Pancreas: Correlation With Secretin-Stimulated Endoscopic Pancreatic Function Test

The purpose of this study was to determine the relationship between the pancreatic enhancement on serial contrast-enhanced MRI (CEMRI) and pancreatic exocrine function using the secretin-stimulated endoscopic pancreatic function test (ePFT). A total of 30 patients with clinical symptoms consistent with chronic pancreatitis underwent CEMRI of the abdomen and ePFT within a 1- to 4-week interval. CEMRI was performed in arterial, early venous, and late venous phases. Secretin ePFT was performed with the measurement of HCO(3) concentration from the duodenal aspirates after secretin stimulation. Contrast enhancement ratio of the arterial phase to early venous phase was measured on CEMRI (SIRa/SIRv). A three-point evaluation system was used for grading the HCO(3) concentration and the enhancement ratio on MRI. For the significance of correlation, kappa statistics was used. Sensitivity and specificity of CEMRI was determined for the diagnosis of early chronic pancreatitis accepting ePFT as a reference. Twenty patients had identical scores on both secretin ePFT and CEMRI. Ten patients revealed discrepancy in scores. Kappa statistics revealed moderate agreement between MRI and ePFT (kappa = 0.44). Sensitivity and specificity values for the diagnosis of pancreatitis were 82% and 57%, respectively. Positive predictive value was 56%, and negative predictive value was 86%. The results of our data indicate that serial CEMRI is an appropriate imaging technique to rule out early chronic pancreatitis. However, secretin-stimulated imaging or ePFT may still be needed for the definite diagnosis of pancreatic exocrine dysfunction.

Clinical applications of 2-D dynamic contrast-enhanced MR subtraction angiography in Neurosurgery — Preliminary results

Objectives Advances in MR hardware performance and imaging techniques have enabled us to perform subsecond frame rate contrast enhanced MR angiographies as a clinical routine, resulting in subtraction angiographies with temporal resolutions similar to those obtained by conventional DSA. In this manuscript, we describe the technique of 2-D dynamic contrast-enhanced MR subtraction angiography and its possible applications in neurosurgery. Methods A T 1 -weighted gradient-echo sequence with short TR and TE values was performed during the intravenous bolus application of MR contrast media. This sequence had a temporal resolution of up to 0.34 seconds per image and was performed in 28 patients harbouring arteriovenous malformations (n=12), dural AV fistulas (n=3), meningiomas (n=8), glomus tympanicum or jugulare tumours (n=3), and haemangioperizytomas (n=2). Arterial digital subtraction angiography (DSA) was performed in all patients and compared to the MRA sequences with respect to differentiation into early, late arterial and venous phases, detectability of normal vessels, detection of the pathology, feeding arteries, capillary blush, early draining veins, and hypervascularisation. Results In all 28 investigated cases and all pathological entities separation of early and late arterial phases, capillary phases, early and late venous phases was possible. The proximal portions of the major cerebral arteries, the capillary blush and the venous anatomy including the tributaries to the large sinuses could be well defined and correlated well with the gold standard, i. e. DSA. Conclusions Dynamic contrast enhanced MR angiography will prove helpful in 1) the follow-up of AVMs since early venous drainage can be visualised; 2) the demonstration of dural AVF, which can be recognised as an early filling of a dural sinus during the early arterial phase; and 3) the characterisation of the degree of vascularisation of brain tumours.

Clinical Applications of 2-D Dynamic Contrast-Enhanced MR Subtraction Angiography in Neurosurgery - Preliminary Results

Advances in MR hardware performance and imaging techniques have enabled us to perform sub-second frame rate contrast enhanced MR angiographies as a clinical routine, resulting in subtraction angiographies with temporal resolutions similar to those obtained by conventional DSA. In this manuscript, we describe the technique of 2-D dynamic contrast-enhanced MR subtraction angiography and its possible applications in neurosurgery. A T (1)-weighted gradient-echo sequence with short TR and TE values was performed during the intravenous bolus application of MR contrast media. This sequence had a temporal resolution of up to 0.34 seconds per image and was performed in 28 patients harbouring arteriovenous malformations (n = 12), dural AV fistulas (n = 3), meningiomas (n = 8), glomus tympanicum or jugulare tumours (n = 3), and haemangioperizytomas (n = 2). Arterial digital subtraction angiography (DSA) was performed in all patients and compared to the MRA sequences with respect to differentiation into early, late arterial and venous phases, detectability of normal vessels, detection of the pathology, feeding arteries, capillary blush, early draining veins, and hyper-vascularization. In all 28 investigated cases and all pathological entities separation of early and late arterial phases, capillary phases, early and late venous phases was possible. The proximal portions of the major cerebral arteries, the capillary blush and the venous anatomy including the tributaries to the large sinuses could be well defined and correlated well with the gold standard, i. e. DSA. Dynamic contrast enhanced MR angiography will prove helpful in 1) the follow-up of AVMs since early venous drainage can be visualised; 2) the demonstration of dural AVF, which can be recognised as an early filling of a dural sinus during the early arterial phase; and 3) the characterisation of the degree of vascularisation of brain tumours.

Differentiation of focal nodular hyperplasia and hepatocellular adenoma by contrast-enhanced ultrasound

Non-invasive differentiation of focal nodular hyperplasia (FNH) and hepatocellular adenoma (HCA) is difficult. The aim of this study was to assess the accuracy of contrast-enhanced phase inversion ultrasound to differentiate between histologically proven FNH and HCA, analysing the arterial and (early) portal venous phase. 32 patients with histological proven FNH (n=24) or HCA (n=8) have been included in this prospective study. Examination technique: Siemens Elegra, phase inversion harmonic imaging (PIHI) with low mechanical index (MI)<0.2-0.3 using SonoVue (BR 1). The contrast enhancing tumour characteristics were evaluated during the hepatic arterial (starting 8-22 s) and early portal venous phase (starting 12-30 s). The image analysis was performed by three examiners. In 23 of 24 patients with FNH the contrast pattern revealed pronounced arterial and (early) portal venous enhancement. Homogeneous enhancement was detected during the hepatic arterial phase in all eight patients with HCA. In contrast to patients with FNH, no enhancement was seen during the portal venous phase. In conclusion, contrast-enhanced phase inversion ultrasound demonstrated pronounced arterial and portal venous enhancement in patients with focal nodular hyperplasia. In contrast, after homogeneous enhancement during hepatic arterial phase, no enhancement during hepatic portal venous phase was detected in patients with hepatocellular adenoma. Therefore, this technique might improve the functional characterization of benign hypervascular focal liver lesions.

Hepatic CT Enhancement: Comparison between Dimeric and Monomeric Nonionic Contrast Agents in Rabbits

Purpose: To determine the hepatic and vascular enhancement profiles with nonionic dimeric, iodixanol, contrast agent in the rabbit and to compare them with nonionic monomeric, ioversol, contrast agent. Materials and Methods: Seven rabbits initially underwent hepatic dynamic CT scan with either iodixanol or ioversol, followed by repeated CT scan with other unused contrast agent with one week interval between scans. Pre and post contrast attenuation values of hepatic parenchyma, aorta and portal vein were measured sequentially. The mean enhancement of the hepatic parenchyma, aorta and portal vein were compared between two agents. The mean peak enhancement and peak enhancement time of the liver, aorta, and portal vein were also compared. Results: The attenuation values of ioversol showed a greater mean hepatic enhancement than iodixanol from 18 seconds to 39 seconds after injection (from late arterial phase to early portal venous phase) with a statistical significance (p

Suspected early or mild chronic pancreatitis: enhancement patterns on gadolinium chelate dynamic MRI. Magnetic resonance imaging.

To assess whether measuring the pattern of pancreatic enhancement on gadolinium chelate dynamic magnetic resonance imaging (MRI) is helpful for diagnosis of suspected early or mild chronic pancreatitis. In this retrospective study, 24 patients with suspected early or mild chronic pancreatitis, classified by imaging criteria of equivocal chronic pancreatitis (ultrasound, computed tomography [CT] or ERCP) grading, had dynamic MRI that included unenhanced, arterial dominant, early venous, and late venous phases of contrast enhancement. Twenty patients without pancreatic diseases also had the dynamic sequence as a control group. The signal intensity was measured at the pancreatic head, body, and tail on all phases, and for each, the signal intensity ratio (SIR, the signal intensity in postcontrast divided by that in precontrast) was calculated. Two radiologists independently reviewed the images of the patients with suspected early or mild chronic pancreatitis for pancreatic morphologic abnormalities without knowing the results of signal intensity measurements. On unenhanced images, there was no significant difference of signal intensity between control and pancreatitis groups (P < 0.05). In the pancreatitis group, but not in the control group, the unenhanced signal intensity of the pancreatic head and body were significantly higher than that of the tail (P < 0.05). In the control group, the greatest enhancement (highest SIR) after injection was in the arterial phase (1.89 +/- 0.31), significantly higher than that in the early venous phase (1.68 +/- 0.17, P < 0.01) and in the late venous phase (1.61 +/- 0.15, P < 0.001). The pancreatitis group, however, had an arterial phase SIR (1.65 +/- 0.23) that was significantly lower than its early venous phase SIR (1.75 +/- 0.22, P < 0.05) and lower than the arterial phase SIR of the control group (P < 0.01). The presence of an SIR less than 1.73 in the arterial phase and/or a delayed peak enhancement after contrast agent administration had a sensitivity and specificity of diagnosing early or mild chronic pancreatitis of 92% and 75%, respectively. This sensitivity was significantly higher than the sensitivity of 50% for diagnosis based on morphologic abnormalities (P < 0.05). Measuring pancreatic signal intensity on gadolinium chelate dynamic MRI is helpful for diagnosing early or mild chronic pancreatitis, especially before apparent pancreatic morphologic or signal intensity changes are present.

Retinal fluorescein contrast arrival time of young patients with the hepatosplenic form of the Schistosomiasis mansoni.

Schistosoma mansoni is responsible for lesions that can alter the hemodinamic of the portal venous circulation, lung arterial and venous sistemic systems. Therefore, hemodinamic changes in the ocular circulation of mansonic schistosomotic patients with portal hypertension and hepatofugal venous blood flow is also probable. The purpose of this study was to determine the fluorescein contrast arrival time at the retina of young patients with the hepatosplenic form of schistosomiasis, clinically and surgically treated. The control group included 36 non schistosomotic patients, mean age of 17.3 years, and the case group was represented by 25 schistosomotic patients, mean age of 18.2 years, who were cared for at The University Hospital (Federal University of Pernambuco, Brazil), from 1990 to 2001. They underwent digital angiofluoresceinography and were evaluated for the contrast arrival time at the early retinal venous phase of the exam. Both groups were ophthalmologically examined at the same hospital (Altino Ventura Foundation, Recife, Brazil), using the same technique. There was retardation of the retinal contrast arrival time equal or more than 70 sec in the eyes of three schistosomotic patients (12%) and in none of the control group, however, the mean contrast arrival time between the two groups were not statistically different. These findings lend support to the hypothesis that there could be a delay of the eye venous blood flow drainage.

Hepatic outflow obstruction: enhancement patterns of the liver on MR angiography

Our purpose was to present the enhancement patterns of the liver on MR angiography in patients with hepatic outflow obstruction. Twenty-three patients with Budd-Chiari syndrome (4 in acute stage and 19 in chronic stage of the disease) were examined with 3D contrast-enhanced MR angiography. During early and late portal venous phase of MR angiography the pattern of parenchymal enhancement was assessed on source images. The enhancement patterns were evaluated under 4 groups as following: (a) central (b) peripheral (c) patchy and (d) homogeneous enhancement. The morphologic changes in the liver (lobar hypertrophy or atrophy, hepatic surface irregularities) were also recorded. In the acute stage global liver enlargement (75%) with caudate hypertrophy (100%) and central enhancement of the liver (75%) were suggestive findings of the hepatic outflow obstruction. The left lobe hypertrophy (53%) associated with the caudate lobe hypertrophy (72%) and irregular surface (26%) were predominant in the chronic stage of the disease. The enhancement patterns seen in chronic disease were variable and reflected the persistent stasis of the portal blood flow (patchy enhancement in 32% of the patients) or the altered hemodynamics of the liver due to the development of subcapsular collaterals (peripheral enhancement in 21% of the patients). Homogeneous enhancement of the liver in Budd-Chiari syndrome may indicate the chronicity of the outflow obstruction (37%) and shows a more stable hepatic perfusion that occurs after the formation of intra and extrahepatic collateral veins. The morphological and perfusional features on multiphase contrast-enhanced MR angiography are valuable in understanding the effects of the hepatic outflow obstruction on the liver parenchyma.

Aging changes of the choroidal dye filling pattern in indocyanine green angiography of normal subjects.

To study the aging changes in the choroid of healthy volunteers with indocyanine green (ICG) angiography. Video ICG angiography with adjunctive computer-assisted image analysis was performed on 35 eyes of 30 healthy volunteers (age range, 21-81 years; mean +/- standard deviation, 50.5 +/- 16.2 years) to observe the aging changes of the choroid. In patients in the second and third decades of life, the subfoveal choroidal arterioles fluoresced initially with subsequent rapid filling of the feeding arterioles and choriocapillaris. The watershed zone was clearly observed. In patients older than age 50, the choroidal vasculature filled more slowly. Eventually, the margin of the watershed zone became blurred. The quantitative analysis showed that the number of choroidal arterioles and the fluorescent intensity in the macular region were reduced with age (P < 0.005). In the early venous phase, hypofluorescent patches seen in all ages increased in size and number and remained with aging. The mean fluorescence intensity was not correlated statistically with age. The features of normal aging patterns of the choroid that we investigated are essential to the interpretation of ICG angiography and may help in understanding the physiologic and pathologic conditions of the choroidal circulations and the choroid and retina themselves.

Gadolinium-infusion magnetic resonance angiogram: A new, noninvasive, and accurate method of preoperative localization of impalpable undescended testes

Purpose: This study investigates the usefulness of Gadolinium (Gd)-infusion magnetic resonance angiogram (MRA) in the detection of impalpable undescended testes. Methods: Magnetic resonance (MRI) examination was performed in 14 patients with 17 impalpable undescended testes (unilateral, n = 11; bilateral, n = 3). MRA was then performed as follows: Gadolinium-DTPA-BMA (Omniscan) at 0.3 mmol/kg body weight was injected intravenously, and dynamic coronal examination using fast spoiled gradient recalled sequences was obtained with images taken at early arterial and delay venous phases. The testis was located by detection of the enhanced pampiniform venous plexus. Results: Of the 17 undescended testes, 11 canalicular hypoplastic testes and 3 intraabdominal testes were detected in both routine MRI and MRA. Three atrophic testes were found in the scrotum by MRA but not detected by routine MRI. The detection rates of impalpable testes by MRI and MRA were 82.4% and 100%, respectively. All imaging findings were confirmed by surgical exploration. There was no morbidity associated with MRA. Conclusion: Gd-infusion MRA is an accurate and sensitive method of preoperative localization of impalpable testes and is superior to conventional MRI in the detection of atrophic testes.

Foot Venous Pressure Measurement in Patients with Varicose Veins

For the evaluation of the severity of varicose veins, foot venous pressure (FVP) measurements utilizing manual calf compression were applied to 52 limbs with varicose veins (group V) and 12 limbs of healthy volunteers (group C). Group V was divided into three groups: group V1 comprising 37 limbs without stasis syndrome, group V2 comprising 9 limbs with pigmentation, and group V3 comprising 6 limbs with preulcerative or ulcerative skin change. First, comparison of manual calf compression with the toe-stand test was performed, wherein a higher reproducibility was revealed for subjects undergoing manual calf compression. Also, a lower ambulatory venous pressure (AVP) and a higher percentage of FVP drop (%drop = [pressure at rest-AVP]/pressure at rest) was obtained with manual calf compression, confirming that this method apparently emptied the calf venous pool. Based on these findings, FVP at four seconds after compression (4SP) and its ratio of increase (4SR% = [4SP - AVP] / [pressure at rest - AVP]) were instituted by use of this method, in addition to AVP, % drop, and 50%/90% of calf-refilling times (RT50/RT90). When the values of the four groups were compared, differences were insignificant between groups C and V1, while significant differences were found between groups V2 (continued on next page) (Abstract continued) or V3 and groups C or V1 in all parameters examined. The number of abnormal limbs that had values beyond the ranges for group C increased from group V1 to group V3. Such abnormalities were best revealed by 4SR% and RT50. In the postoperative FVP measurements of 17 limbs, all parameters of all limbs were restored to the normal range, except for % drop of 4 limbs, and 4SR%, RT50, and RT90 of 1 limb. Based on these findings, all parameters examined in this study were considered to be sufficient for the evaluation of the severity of varicose veins. In particular, 4SR% and RT50, which reflect venous reflux in the early venous refilling phase, were the best diagnostic parameters. The usefulness of FVP measurement utilizing manual calf compression and the significance of the early phase of venous refilling in the varicose veins limbs were confirmed.

Choroidal filling patterns in sickle cell patients

Two choroidal occlusion syndromes are recognized: the triangular syndrome after ciliary artery occlusions, and the multifocal lesions associated with choroidal arteriolar occlusions. Both syndromes have been described in sickle cell patients. We compared early venous phase fluorescein angiograms of 51 Hb SS or Hb SC patients and 15 age and race matched controls, to test whether geographic early filling defects of the choriocapillaris might be more common in these rheologically impaired patients. We found no difference in occurrence of these filling defects between the two groups, and conclude that choriocapillaris filling defects in the early venous phase only are not pathological.

Medical pattern recognition pitfalls in a clinical setting: renal cell carcinoma survival prediction.

Three studies to predict renal cell carcinoma patient survival from tumor textures found in digitized early venous phase arteriograms were successful individually. However, when the three methods were compared, they were not consistent and no single method was clinically useful. The first study predicted 5 yr survival of 37 patients with 87% accuracy. The second study added 29 patients to the data base; the poor survival of the 21 patients who died within 5 yr of diagnosis was predicted with 80.9% accuracy. When 27 of these cases were redigitized with a laser scanner, average survival prediction accuracy was 78%. In these studies, digitization hardware, radiographic technique, normalization methods, window selection, and contrast medium distribution all contributed to differences in the statistics separating poor from good patient survival.

Vasoactive effects of buflomedil assessed by fluorescein angiography.

The vascular effects of buflomedil were studied by retinal fluorescein angiography carried out before (control) and after the intravenous infusion of buflomedil (250 mg in 15 min) in patients with different pathological conditions of retinal circulation. Sequential photograms, taken every 36 sec from the 10th sec after the start of the fluorescein injection, showed differences in circulation times between the two sequences in 9 (75%) out of 12 patients. Arterial phase and early venous phase differences were evident, with a mean delay after buflomedil of 3.24 +/- 2.35 sec and 3.22 +/- 2.58 sec, respectively. Significant differences were observed in the time to venous filling completion, with a mean delay of 6.99 +/- 5.19 sec after buflomedil. It is considered that the delay in the retinal arterial phase is related to the arm-retina circulation time, and the slowing down in venous filling is related to a lengthened circulation in the capillary bed.