Magnetic resonance imaging of anal cancer: tumor characteristics and early prediction of treatment outcome.

PURPOSE: To analyze tumor characteristics derived from pelvic magnetic resonance imaging (MRI) of patients with squamous cell carcinoma of the anus (SCCA) before and during chemoradiotherapy (CRT), and to compare the changes in these characteristics between scans of responders vs. nonresponders to CRT. METHODS: We included 52 patients with a pelvic 3T MRI scan prior to CRT (baseline scan); 39 of these patients received an additional scan during week 2 of CRT (second scan). Volume, diameter, extramural tumor depth (EMTD), and external anal sphincter infiltration (EASI) of the tumor were assessed. Mean, kurtosis, skewness, standard deviation (SD), and entropy values were extracted from apparent diffusion coefficient (ADC) histograms. The main outcome was locoregional treatment failure. Correlations were evaluated with Wilcoxon's signed rank-sum test and Pearson's correlation coefficient, quantile regression, univariate logistic regression, and area under the ROC curve (AUC) analyses. RESULTS: In isolated analyses of the baseline and second MRI scans, none of the characteristics were associated with outcome. Comparison between the scans showed significant changes in several characteristics: volume, diameter, EMTD, and ADC skewness decreased in the second scan, although the mean ADC increased. Small decreases in volume and diameter were associated with treatment failure, and these variables had the highest AUC values (0.73 and 0.76, respectively) among the analyzed characteristics. CONCLUSION: Changes in tumor volume and diameter in an early scan during CRT could represent easily assessable imaging-based biomarkers to eliminate the need for analysis of more complex MRI characteristics.

Efficacy of plain radiography and computer tomography in localizing the site of pelvic arterial bleeding in trauma patients.

BACKGROUND: Immediate angiography is warranted in pelvic trauma patients with suspected arterial injury (AI) in order to stop ongoing bleeding. Prior to angiography, plain pelvic radiography (PPR) and abdominopelvic computer tomography (CT) are performed to identify fracture and hematoma sites. PURPOSE: To investigate if PPR and CT can identify the location of AI in trauma patients undergoing angiography. MATERIAL AND METHODS: 95 patients with pelvic fractures on PPR (29 women, 66 men), at a mean age of 44 (9-92) years, underwent pelvic angiography for suspected AI. Fifty-six of them underwent CT additionally. Right and left anterior and posterior fractures on PPR were registered, and fracture displacement was recorded for each quadrant. Arterial blush on CT was registered, and the size of the hematoma in each region was measured in cm(2). AIs were registered for anterior and posterior segments of both internal iliac arteries. Presence of fractures, arterial blush, and hematomas were correlated with AI. RESULTS: Presence of fracture in the corresponding skeletal segment on PPR showed sensitivity and specificity of 0.86 and 0.58 posteriorly, and 0.87 and 0.44 anteriorly. The area under the curve (AUC) was 0.77 and 0.69, respectively. Fracture displacement on PPR >0.9 cm posteriorly and >1.9 cm anteriorly revealed specificity of 0.84. Sensitivities of arterial blush and hematoma on CT were 0.38 and 0.82 posteriorly, and 0.24 and 0.82 anteriorly. The specificities were 0.96 and 0.58 posteriorly, and 0.79 and 0.53 anteriorly, respectively. For hematomas, the AUC was 0.79 posteriorly and 0.75 anteriorly. Size of hematoma >22 cm(2) posteriorly and >29 cm(2) anteriorly revealed specificity of 0.85 and 0.86, respectively. CONCLUSION: CT findings of arterial blush and hematoma predicted site of arterial bleeding on pelvic angiography. Also, PPR predicted the site of bleeding using location of fracture and size of displacement. In the hemodynamically unstable patient, PPR may contribute equally to effective assessment of injured arteries.

Intraparenchymal Doppler ultrasound after proximal embolization of the splenic artery in trauma patients.

We studied the changes in proximal embolization of the splenic artery to the intraparenchymal blood flow with Doppler ultrasound. Seventeen trauma patients with spleen injury OIS grade 2-5 underwent embolization of the splenic artery. Peak-systolic velocity (PSV) and end-diastolic velocity (EDV) were measured in intrasplenic arteries initially 1 day after embolization, at early follow-up after 7 days, at intermediate follow-up after 10 weeks, and at late follow-up after 10 months. Resistance index (RI), systolic/diastolic ratio (S/D ratio), acceleration (AC), and acceleration time (AT) were calculated. The results were compared to values from 17 volunteers. RI increased from 0.39 initially to 0.49 (P = 0.002) at intermediate and to 0.52 (P < 0.001) at late follow-up. S/D ratio increased from 1.68 initially to 1.99 (P = 0.002) and to 2.10 (P < 0.001) at intermediate and late follow-up, respectively. Follow-up results of RI and S/D ratio differed significantly from the reference group. AC increased from 1.06 m/s(2) initially to 1.89 m/s(2) at late follow-up (P = 0.01). AC at late follow-up was not different from reference group (2.33 m/s(2)). In conclusion, Doppler ultrasound is a useful tool in the evaluation of improvable intraparenchymal blood flow over time after central splenic artery embolization.

Cardiac accumulation of bone marrow mononuclear progenitor cells after intracoronary or intravenous injection in pigs subjected to acute myocardial infarction with subsequent reperfusion.

OBJECTIVE: The purpose of the present study was to compare the efficacy of intracoronary and intravenous injection of autologous progenitor cells for homing to the acutely infarcted but reperfused myocardium in pigs. METHODS: Myocardial infarction was induced in 11 anesthetized pigs by 60-min balloon inflation in the mid LAD. After balloon deflation, reperfusion was verified and autologous CD31(+) progenitor cells, or bone marrow mononuclear cells, labeled with PKH67, were injected either intracoronarily (n=6) or intravenously (n=3). By autopsy, 4-5 days after induction of infarction, tissue from the heart and other organs was obtained for fluorescence microscopy. RESULTS: In the heart, PKH(+) cells were detected throughout the reperfused infarcted myocardium, and the number of PKH(+) cells was significantly higher after intracoronary than after intravenous injection (3.2+/-0.55 vs. 0.33+/-0.17 cells/high-power field/10(6) cells injected, P=.01). Few PKH(+) cells were detected in the spleen, lung, mesenteric lymph node, and bone marrow. In an additional animal with a coil placed in the mid LAD, progenitor cells were not detected in the infarcted myocardium or in the normal myocardium. CONCLUSION: Autologous mononuclear and CD31(+) cells from bone marrow accumulated in the infarcted myocardium when injected intracoronarily or intravenously after established reperfusion, and the accumulation of cells was significantly greater after intracoronary injection than after intravenous injection. Accumulation of PKH(+) cells did not appear in the normal myocardium or in the nonreperfused infarcted myocardium. PKH(+) cells were detected in spleen, lung, and bone marrow but to a lesser degree than in the infarcted myocardium.

Osmolarity: a decisive parameter of bowel agents in intestinal magnetic resonance imaging.

The aim was to evaluate the importance of the osmolarity of different oral agents for bowel distension and the level of related adverse events. The longitudinal design included the exposition of different oral MR agents on two separate occasions. Four groups of volunteers were randomly given 350 ml gastrografin of three different concentrations and water. On the second occasion they received mannitol, iohexol or iodixanol with equivalent osmolarities, but the control group (water) received mannitol. We recorded the outcomes as the degree of bowel distension determined as the mean bowel section area and the total level of discomfort recorded from a visual analogue scale (VAS). The statistical analysis included scatter plots with the best-fitted line with linear regression to study the association between osmolarity and section area and the association between osmolarity and adverse events. A dose-response association was found between increasing osmolarity levels and bowel area in square centimeters (P = 0.00001). A similar dose-response association existed between increasing levels of osmolarity and adverse events (P = 0.001). Osmolarity appears to be more important for bowel distension than the physico-chemical characteristics of the nonabsorbable oral agents. The optimum osmolarity level is determined by the patient's tolerance of the adverse events.

MR imaging of the small bowel with increasing concentrations of an oral osmotic agent.

The aim of this study was to assess the quality of MR imaging and level of adverse effects with increasing concentrations of gastrografin. This is a prospective study with 24 healthy volunteers which were randomised into four groups receiving 50%, 25%, 10% and 0% gastrografin. The endpoint was bowel image quality based on distension, signal homogeneity and wall delineation evaluated by three independent radiologists, and the maximum bowel diameter at three different levels. The subjects also scored any adverse events on a 1-5 scale. The interradiologist agreement was relatively good, with kappa values varying between 0.81 and 0.41. Improved bowel distension and image quality were achieved with increasing concentrations. But significant dose-response effects were found between increasing osmolalities and the bowel diameters and also versus the score of adverse events. The most frequent adverse reactions were diarrhea, nausea and lack of palatability. There is a gradient relationship between increasing osmolality of gastrografin and improved image quality and the score of adverse effects. The optimum concentration of gastrografin is dependent of the tolerance of the adverse events.