Multi-slice CT for Visualization of Acute Pulmonary Embolism: Single Breath-hold Subtraction Technique

PURPOSE

The purpose of our preliminary animal study was to evaluate the feasibility of a new subtraction technique for visualization of perfusion defects within the lung parenchyma in segmental and subsegmental pulmonary embolism (PE).

MATERIALS AND METHODS

In three healthy pigs, PE were artificially induced by fresh human clot material. Within a single breath-hold, CT angiography (CTA) was performed on a 16-slice multi-slice CT scanner (SOMATOM Sensation 16; Siemens, Forchheim, Germany) before and after intravenous application of 80 mL of contrast-medium, followed by a saline chaser. Scan parameters were 120 kV and 100 mAs (eff.), using a collimation of 16 x 1.5 mm and a table speed/rot. of 36 mm (pitch: 1.5; rotation time: 0.5 s). A new 3D subtraction technique was developed, which is based on automated segmentation, non-linear spatial filtering and non-rigid registration. Data were analysed using a color-encoded "compound view" of parenchymal enhancement and CTA information displayed in axial, coronal and sagittal orientation.

RESULTS

Subtraction was technically feasible in all three data sets. The mean scan time for each series was 4.7 s, interscan delay was 14.7 s, respectively. Therefore, an average breath-hold of approximately 24 s was required for the overall scanning procedure. Downstream of occluded segmental and subsegmental arteries, perfusion defects were clearly assessable, showing lower or missing enhancement compared to normally perfused lung parenchyma. In all pigs, additional peripheral areas with triangular shaped perfusion defects were delineated, considered typical for PE.

CONCLUSIONS

Our initial results from the animal model studied show that perfusion imaging of PE is feasible within a single breath-hold. It allows a comprehensive assessment of perfusion deficits as the direct proof of a pulmonary embolus, can be combined with an indirect visual quantification of the density changes in the adjacent lung tissue.

Mehrschicht-Spiral-CT (MSCT) in der Kinderradiologie: Dosisreduktion bei der Untersuchung von Thorax und Abdomen

The advent of multislice spiral CT (MSCT) technique has led to new aspects of dose reduction, especially for the dedicated use of MSCT in children. Optimizing pediatric MSCT protocols according to the clinical problem allows reduction of radiation exposure to a minimum without loss of diagnostic quality. The different parameters that influence the degree of dose reduction, like tube current-time product (mAs), tube voltage (kV), collimation and pitch, are discussed in context with previously published data and our own experience in nearly 200 pediatric CT examinations. In our department, the effective mAs is calculated for a pediatric chest MSCT by multiplication of the body weight in kilogram with a factor of 1 to 1.5 and for a pediatric abdominal MSCT by multiplication with a factor of 2 to 2.5. To calculate the equivalent effective dose for a contrast media-enhanced 80 kV protocol, the effective mAs of the 120 kV protocol can be multiplied by 2.7. A factor less than 2.7 means further dose reduction. Compared to the radiation exposure with a standard adult protocol, the effective dose in a pediatric thoracic MSCT could be reduced by up to 92 % in neonates, 89 % in toddlers and 80 % in school children. In abdominal MSCT, the effective dose could be reduced by up to 90 % in neonates, 89 % in toddlers and 83 % in school children. Using an adequate MSCT technique in children by adjusting the CT scanning parameters to the clinical question and body weight of the examined child enables a significant reduction of radiation exposure in comparison to standard MSCT protocols.

Quantitative and qualitative assessment of left ventricular volume with ECG-gated multislice spiral CT: value of different image reconstruction algorithms in comparison to MRI

PURPOSE

To assess the value of different image reconstruction algorithms for assessment of the left ventricular function using retrospectively ECG-gated multislice spiral computed tomography (MSCT) of the heart.

MATERIAL AND METHODS

MSCT and cine magnetic resonance (MR) imaging of the heart were performed in 15 patients. For MSCT, standard and multisegmental image reconstruction with improved temporal resolution were used. Standardized multiplanar reformats in the short axis and long axis views were reconstructed from MSCT data. End-systolic (ESV) and end-diastolic volume (EDV), stroke volume (SV), ejection fraction (EF) and myocardial mass (MM) were calculated. Left ventricular wall motion was assessed qualitatively.

RESULTS

MSCT was in acceptable agreement with MR imaging for quantification of the ventricular function. According to the Bland-Altman approach the mean differences for the left ventricular volumes (ESV, EDV, SV) ranged from -9.6 ml to 3.1 ml with standard image reconstruction and from -0.6 ml to 1.9 ml utilizing multisegmental image reconstruction with limits of agreement ranging from -26.6 ml to 12.5 ml and -15.6 ml to 15.0 ml, respectively. Applying the multisegmental image reconstruction algorithm, a significantly improved agreement with the MR data was found for EDV, SV and EF. For wall motion analysis, standard image reconstruction showed a significant difference to MR imaging with a correspondence in 83.75% of the 240 assessed segments, while multisegmental image reconstruction agreed with MR imaging in 92.5% of the segments.

CONCLUSION

Multisegmental image reconstruction improves the quantitative assessment of left ventricular function when compared to standard image reconstruction. Multisegmental image reconstruction allows qualitative wall motion analysis.

Wertigkeit der kardialen Mehrschicht-Spiral-CT zur Beurteilung der degenerativen Aortenklappenstenose: Vergleich zur Echokardiographie

OBJECTIVE

To non-invasively assess the severity of aortic valve stenosis (AS) by the determination of aortic valve calcification (AVC) using multislice spiral computed tomography (MSCT).

MATERIALS AND METHODS

Forty-one consecutive patients (17 male, 24 female, mean age 71.0 +/- 7.9 years) with a history of AS and an aortic valve area < or = 2 cm (2) underwent retrospectively ECG-gated 4-slice MSCT and echocardiography. The AVCs were quantitatively assessed using the score described by Agatston as well as by calculating the calcium mass. The echocardiographically determined aortic valve area (AVA) and the severity of AS according to the ACC/AHA guidelines were compared to the degree of aortic valve calcifications. Pearson's correlation coefficient, cut-off values, kappa test and F-test with post hoc Bonferroni t-tests were calculated.

RESULTS

Calcium scores were significantly higher in patients with severe AS, when compared to mild or moderate AS (p < 0.001). In patients suffering from severe AS, the mean Agatston score was 4125.5 +/- 1168.9 (calcium mass 904.1 +/- 263.3) while in patients with moderate and mild AS the corresponding values were 1596.3 +/- 987.0 (319.1 +/- 208.3) and 785.9 +/- 390.1 (149.1 +/- 90.2), respectively. Pearson's correlation coefficients were r = - 0.75 for the Agatston score and r = - 0.72 for the calcium mass. There was a moderate agreement between severity of AS according to the ACC/AHA guidelines and the degree of AS determined from AVC scores with kappa = 0.6091 and kappa = 0.6985, respectively.

CONCLUSION

Severe AS may be differentiated from moderate or mild AS using cardiac MSCT. Extensive calcifications of the aortic valve presenting with an Agatston-Score > or = 2824 (calcium mass > or = 611) indicate a severe AS and should be taken as an indication for further diagnostic workup.

Verfeinerung der zytologischen Diagnostik von CT-gesteuerten Feinnadelaspirationsbiopsien durch histologische Zusatzuntersuchung Formalin-fixierter Blutkoagel

PURPOSE

To evaluate formalin-fixed blood-clot for diagnostic efficacy and safety in percutaneous CT-guided fine needle aspiration biopsy (FNAB) in addition to cytopathology.

MATERIALS AND METHODS

61 CT-guided FNABs were entered into a prospective trial. FNAB was performed in 22 female and 39 male patients (32 - 83 years, m = 62.8 +/- 10.9 y) using 20- (n = 7) and 22-gauge (n = 54) needles. After correct placement within the tumor, material for cytopathological work-up was harvested. Additional material was sampled by needling (stirring and aspirating) within the lesion. These blood-clots were fixed in buffered formalin and delivered for histopathological work-up along with the smears. All results were correlated with the clinical course of the patient; cases of negative biopsies underwent surgery or were followed clinically for a minimum of 13 months (m = 14.8 +/- 1.4 months). Finally, all puncture-related side-effects and complications were analyzed and compared with data from 108 transthoracic FNABs performed over a two-year period before using this technique.

RESULTS

Tissue samples adequate for pathological diagnosis were obtained in 93.4 % (57 of 61 patients). Malignant cells were found in 50 samples, and 7 biopsy results were negative. The relevant pathological changes were depicted on both smear and blood-clot in 47 cases (82.4 %). In the remaining patients, the diagnoses were exclusively established by cytopathology (n = 6) and histopathology (n = 4). Altogether, additional information was obtained in 20 of the 57 FNABs (35.1 %). In the 47 cases with relevant findings obtained with both methods, the routine stain already led to the correct diagnosis in 9 cases by showing cohesive layers of malignant cells, and immunocytochemistry allowed subclassification of the biopsy material in additional 7 cases. Pneumothorax occurred in 12 of 38 transpleural punctures (31.6 %), hemoptysis in one patient (2.6 %). Four pneumothoraces required further treatment (3 thoracocenteses, 1 chest tube placement). Hemorrhagic complications requiring further treatment were not seen, but two minor hematomas were observed in the 19 abdominal punctures. The overall rate for pneumothorax was 38.9 % (42 of 108 interventions) for the FNAB performed in standard technique.

CONCLUSION

Histological work-up of sampled blood-clots yielded a higher accuracy of CT-guided FNABs. Additional immunocytochemical studies allowed subclassification of tumor material. The complication rate of this technique was not higher than for FNAB alone.

Individually weight-adapted examination protocol in retrospectively ECG-gated MSCT of the heart

The standard protocol in multislice spiral CT (MSCT) angiography for coronary arteries with fixed tube current-time settings leads to an overexposure and thus to an unnecessary high radiation dose in patients with lower weight when compared to heavier patients. The purpose of this study was to estimate the effect of reducing the radiation dose by adapting the tube current-time settings individually. Fifty patients underwent retrospectively ECG-gated MSCT of the heart. In 25 patients (group A1) a standard protocol with constant tube current-time settings was used (4 x 1-mm collimation, 120 kV, 400 mAs(eff)). Subsequently, artificial image noise was added to the data of these patients simulating a directive for weight-adapted tube current-time settings (group A2). In the other 25 patients (group B) an alternative protocol with individually weight-adapted tube current-time settings was applied. The data of all groups were evaluated by a regression analysis. The image quality was assessed objectively by measuring the CT attenuation in standardised regions of interest and subjectively by three radiologists using a five-point scoring system in a consensus reading. Applying the weight-adapted tube current-time settings the effective radiation dose was reduced by 17.9% for men and 26.3% for women. The standard protocol leads to an overexposure in light patients as seen in the plot of noise vs weight (slope 0.16+/-0.07 HU/kg). By applying the weight-adapted tube current-time settings a weight-independent, constant image noise is achieved (slope 0.04+/-0.1 HU/kg). Diagnostic image quality was preserved in all patients. Individually weight-adapted tube current-time settings allow for a substantial dose reduction when performing retrospectively ECG-gated MSCT angiography for coronary arteries without impairment of diagnostic image quality.

Detection of coronary calcifications: feasibility of dose reduction with a body weight-adapted examination protocol

OBJECTIVE

The purpose of this study was to assess the applicability of individual body weight-adapted tube current time settings in multidetector CT for detection of coronary calcifications and to evaluate the effect of reducing the radiation dose on the coronary calcium score. SUBJECTS AND METHODS. One hundred patients underwent retrospectively ECG-gated MDCT for detection of coronary calcifications. First, fixed tube current time settings were used in 50 patients. Second, image noise corresponding to body weight-adapted tube current time settings was added to these images. Finally, body weight-adapted tube current time settings were applied to another 50 patients. For each patient group, the radiation dose was calculated. Coronary calcium scores were compared for the patient groups with the fixed tube current time settings with and without artificially added image noise. In all image series, image noise was assessed by a region-of-interest methodology. Image noise was analyzed using a regression analysis.

RESULTS

The effective radiation dose was reduced by 11.6% for men and 24.8% for women using the body weight-adapted tube current time settings. There were no statistically significant changes in the coronary calcium score after the addition of artificial image noise (p = 0.84). Adaptation of the tube current time settings did not lead to a relevant increase in image noise. The radiation doses for the plotted noise-to-body weight (slope, 0.081) and noise-to-body mass index (slope, 0.378) ratios for the standard protocol proved relatively high for patients of lower weight. An improved noise-to-body weight (slope, 0.054) and noise-to-body mass index (slope, 0.190) ratio was achieved by application of the body weight-adapted tube current time settings, resulting in nearly constant image noise related to body weight.

CONCLUSION

Individual body weight-adapted current time settings are applicable for coronary calcium scoring without a change of the coronary calcium score or relevant increase of the image noise.

Spatial domain filtering for fast modification of the tradeoff between image sharpness and pixel noise in computed tomography

In computed tomography (CT), selection of a convolution kernel determines the tradeoff between image sharpness and pixel noise. For certain clinical applications it is desirable to have two or more sets of images with different settings. So far, this typically requires reconstruction of several sets of images. We present an alternative approach using default reconstruction of sharp images and online filtering in the spatial domain allowing modification of the sharpness-noise tradeoff in real time. A suitable smoothing filter function in the frequency domain is the ratio of smooth and original (sharp) kernel. Efficient implementation can be achieved by a Fourier transform of this ratio to the spatial domain. Separating the two-dimensional spatial filtering into two subsequent one-dimensional filtering stages in the x and y directions using a Gaussian approximation for the convolution kernel further reduces computational complexity. Due to efficient implementation, interactive modification of the filter settings becomes possible, which can completely replace the variety of different reconstruction kernels.

Value of 3D-volume rendering in the assessment of coronary arteries with retrospectively ECG-gated multislice spiral CT

PURPOSE

To assess the diagnostic value and measurement precision of 3D volume rendering technique (3D-VRT) from retrospectively ECG-gated multislice spiral CT (MSCT) data sets for imaging of the coronary arteries.

MATERIAL AND METHODS

In 35 patients, retrospectively ECG-gated MSCT of the heart using a four detector row MSCT scanner with a standardized examination protocol was performed as well as quantitative X-ray coronary angiography (QCA). The MSCT data was assessed on segmental basis using 3D-VRT exclusively. The coronary artery diameters were measured at the origin of each main coronary branch and 1 cm, 3 cm and 5 cm distally. The minimum, maximum and mean diameters were determined from MSCT angiography and compared to QCA.

RESULTS

A total of 353 of 525 (67.2%) coronary artery segments were assessable by MSCT angiography. The proximal segments were more often assessable when compared to the distal segments. Stenoses were detected with a sensitivity of 82.6% and a specificity of 92.8%. According to the Bland-Altman method the mean differences between QCA and MSCT ranged from -0.55 to 1.07 mm with limits of agreement from -2.2 mm to -2.7 mm.

CONCLUSION

When compared to QCA, the ability of 3D-VRT to quantitatively assess coronary artery diameters and coronary artery stenoses is insufficient for clinical purposes.

Aneurysmal bone cyst: value of MR imaging and conventional radiography

Our objective was to investigate the accuracy and the diagnostic value of different imaging features of primary aneurysmal bone cysts (ABC) in conventional radiographs and MRI. Conventional radiographs and MR images of 34 patients with a suspected aneurysmal bone cyst were reevaluated by six independent radiologists in a blinded fashion. Morphological features, MR signal characteristics, and contrast enhancement patterns were assessed. Diagnoses were correlated with histology. Sensitivity and specificity of the different imaging findings for each imaging technique were calculated. In 24 patients ABC was histologically proven. In 10 cases diagnoses other than ABC were established. Conventional radiography and MRI each demonstrated a sensitivity of 76.4 and 77.8% with a specificity of 55.0 and 66.7%, respectively. With combined use of both imaging modalities sensitivity and specificity increased to 82.6 and 70%, respectively. The ABC was significantly more often seen as circumscribed lesion with bone expansion, lobulation, septa, and pathological cortical bone reaction than those cases with different diagnoses. Septal contrast enhancement proved to be a useful MR imaging finding in suspected ABC. The combined use of conventional radiographs and MRI revealed the highest sensitivity in the diagnosis of ABC in relation to histology. Magnetic resonance imaging provides improved diagnostic specificity and valuable information on soft tissue changes.

[Quantification of cardiac function with multislice spiral CT using retrospective EKG-gating: comparison with MRI]

PURPOSE

To quantify left ventricular function derived from retrospectively ECG-gated multislice spiral CT (MSCT) data sets in comparison to MRI.

MATERIALS AND METHODS

In 16 patients (14 males, 2 females, mean age 56.8 +/- 11.5 years), retrospectively ECG-gated MSCT angiography of the coronary arteries and breath-hold steady state free precession cine MRI were performed. From MSCT data-sets, 20 axial image series were reconstructed every 5 % of the RR interval. Multiplanar images were reformatted in the short axis orientation from axial images. End-systolic and end-diastolic images were selected. From these images end-systolic volume (ESV), end-diastolic volume (EDV) and stroke volume (SV) as well as the ejection fraction (EF) and myocardial mass (MM) were determined using the Simpson's method and compared with MRI. Furthermore, image quality was assessed for both imaging modalities using a four point grading scale.

RESULTS

All parameters were found to have an excellent correlation between MSCT and MRI data (Pearson's correlation coefficient 0.95 - 0.99), without clinically relevant differences between both modalities. On average, the difference between both methods was 0.5 ml for ESV, 0.8 ml for EDV, 1.3 ml for SV, 0.9 % for EF and 2.3 g for MM. Image quality was slightly better for MRI (1.5 +/- 0.65) than for MSCT (1.64 +/- 0.74).

CONCLUSION

Retrospectively ECG-gated MSCT angiography can not only visualize the coronary arteries but also enables precise quantification of the left ventricular function from the same MSCT data set.

[Accuracy of fluoroscopically navigated drilling procedures at the hip]

AIM

Many orthopaedic procedures require an accurate drilling in bone. The outcome is frequently dependent on the geometric accuracy of this surgical step. The precision of such a procedure can be improved with the help of fluoroscopic navigation. Reliability, accuracy and benefit of this new method for the patient, as well as for the surgical staff, need to be analysed.

METHOD

In a standardised in vitro trial, the drilling of a 5 mm spherical lesion implanted in an artificial femoral head was performed using a navigated drill-guide and a navigated drill. In groups A and B, the distance of the tip of the drill to the center of the lesion was analysed in a 3D CT-generated model and in macroscopic cross section. Additionally, in group B the actual direction of the drill canal was measured.

RESULTS

The mean distance in group A was measured to be 1 mm, with all results ranging between 0 and 2.5 mm. In group B the planned direction of the canal was reproduced with a deviation of 0 degrees to 7 degrees, the target only being missed by a mean distance of 2.5 mm and a maximum of 3.5 mm. Compared to the macroscopic and 3D-CT findings, the correlation of the data calculated by the navigation system was accurate up to a difference of 4 degrees or 2 mm.

CONCLUSION

The fluoroscopically assisted freehand navigation used during the drilling of bone led to a high accuracy of three-dimensional tip placement while reducing radiation exposure to a minimum. It represents a promising and efficient application for a variety of procedures in orthopaedic surgery.

[CT-angiography of the carotid artery: First results with a novel 16-slice-spiral-CT scanner]

PURPOSE

To evaluate a novel multislice CT system (16-slice-spiral-CT scanner) for the diagnosis of carotid artery stenosis.

MATERIAL AND METHODS

Five patients with symptomatic atherosclerotic disease of the carotid arteries were examined with a 16- slice-spiral-CT scanner. Collimation was 16 x 0.75 mm, table speed 36 mm/s (pitch of 1.5), rotation time 0.5 s, tube current was 160 eff.mAs at 120 kV. 60 ml of contrast material were injected with a power injector followed by a saline flush. The start delay was measured with test bolus method (20 ml CM). Interactive multiplanar reformation (iMPR) and thin slab MIP as well as volume rendering were used for image evaluation and presentation.

RESULTS

Scan time was 9 s for a range of 300 mm. This allowed imaging the whole length of the carotid artery (aortic arch to circle of Willis) in a true arterial phase. Pulsation artefacts did not impair the evaluation of the vessels at the level of the aortic arch. Overall image quality of both "source images" and 3D-reconstructions was excellent, due to a reduced voxel size of 0.03 mm (3). Image evaluation and postprocessing (iMPR, MIP) was done within 15 min. iMPR was highly accurate for demonstrating plaque morphology and determining the percentage of the stenosis.

CONCLUSION

For the first time, true arterial phase images of the entire carotid artery with high spatial resolution could be acquired using a 16-slice-spiral-CT scanner. This method offers the potential to replace catheter angiography in the evaluation of carotid artery stenosis.

Pre-transplant positron emission tomography (PET) using fluorine-18-fluoro-deoxyglucose (FDG) predicts outcome in patients treated with high-dose chemotherapy and autologous stem cell transplantation for non-Hodgkin's lymphoma

We investigated the predictive value of sequential FDG PET before and after high-dose chemotherapy (HDT) and autologous stem cell transplantation (ASCT) in 24 patients suffering from non-Hodgkin's lymphoma (NHL). FDG PET was performed at baseline, after three cycles of induction therapy, before and after HDT with ASCT. Response assessment from sequential PET scans using standardized uptake values (SUV) was available in 22 patients at the time of transplantation. Partial metabolic response (PMR) was defined as a >25% decrease of SUV between successive PET scans [corrected]. Six of seven patients who did not achieve a PMR after complete induction therapy developed lymphoma progression, while 10 of 15 patients with complete metabolic response (CMR) or PMR remained in continuous remission. Four of seven patients with less than PMR after induction therapy died vs two of 15 patients with CMR/PMR. Median progression-free and overall survival of patients with less than PMR after HDT and ASCT was 9 and 29 months, respectively. In contrast, neither conventional re-staging nor the International Prognostic Index were predictive. These data suggest that sequential quantitative PET imaging does enlarge the concept of chemosensitivity used to select patients with high-risk NHL for HDT and ASCT or to route them to alternative treatments.

Multidetector CT of the spine in multiple myeloma: comparison with MR imaging and radiography

OBJECTIVE

The purpose of this study was to compare multidetector CT (MDCT) of the thoracic and lumbar segments of the spine with MR imaging and conventional radiography for bone lesion detection and for evaluating the risk of vertebral fracture in multiple myeloma.

SUBJECTS AND METHODS

Eighteen patients with multiple myeloma stage III (according to the criteria of Durie and Salmon) underwent MDCT, conventional radiography, and MR imaging of the lumbar and thoracic spine. MDCT was performed using a standard protocol with no contrast material. Source images were reconstructed using an effective slice thickness of 3 mm with an overlapping reconstruction increment (0.8 mm). Secondary coronal and sagittal multiplanar reformations were exclusively used for establishing the diagnosis. Findings were compared with those of MR imaging and conventional radiography.

RESULTS

In all patients, coronal and sagittal multiplanar reformations depicted the extent of osseous destruction and provided detailed information about osseous infiltration and potential bone instability. Compared with conventional radiography, an additional 24 affected vertebrae, 15 additional vertebral fractures, and six vertebrae at further risk of fracture were detected on MDCT. Compared with MR imaging, three additional endangered vertebrae were detected on MDCT. MR imaging alone would have lead to an understaging of five (27.8%) of 18 patients. Using combined radiography and MR imaging, disease in three (16.7%) of 18 patients would have been understaged.

CONCLUSION

MDCT seems to be preferable to conventional radiography in evaluating bone destruction in multiple myeloma. In combination with MR imaging, detailed information for staging these tumors is obtained. For the initial staging in patients with multiple myeloma, MDCT in combination with MR imaging seems to be the method of choice.

[Visualization of coronary arteries in CT as assessed by a new 16 slice technology and reduced gantry rotation time: first experiences]

PURPOSE

First evaluation of image quality of a new 16-slice multidetector-row computed tomography (MDCT) for the assessment of coronary artery disease and lesion detection of the coronary arteries.

MATERIALS AND METHODS

On a newly developed 16-slice CT scanner (SOMATOM Sensation 16, Siemens, Forchheim, Germany) a calcium score as well as a contrast-enhanced CT angiography (CTA) were performed on 4 patients with retrospective ECG-gating and a gantry rotation time of 420 ms to exclude or follow-up coronary heart disease. CTA was performed after injecting 120 ml contrast media intravenously. After medication with a ss-Blocker, the heart rate was between 55 and 67 bpm.

RESULTS

The scan time for calcium score was 12 s, for CTA 18 s (scan range 15 and 12 cm, respectively). Volume score was between 0 and 256.4. In the CT angiography the entire coronary tree could be visualized in all patients up to the very distal subsegmental branches. In two patients a complete occlusion of the RCA and the LAD were depicted, respectively. In one of these patients, a large aneurysm of the left anterior ventricular wall was also delineated.

CONCLUSION

Considering our first experiences with the new 16-slice technology, an excellent visualization of the entire coronary tree including the very distal and side branches due to substantially increased spatial resolution seems to be achievable. In these patients the acquired image quality raises the hope for improved, non-invasive cardiac diagnostics. In larger studies, the clinical impact of this new technology needs to be further investigated.

[A differentiated approach to the diagnosis of pulmonary embolism and deep venous thrombosis using multi-slice CT]

PURPOSE

To establish a differentiated protocol for multi-slice CT (MSCT) examinations in cases of clinically suspected pulmonary embolism (PE) using pulmonary CT-angiography (CTA) and indirect CT-phlebography (CTP).

MATERIALS AND METHODS

161 patients with suspected PE were examined using an MSCT (SOMATOM Volume Zoom; Siemens, Forchheim, Germany). After intravenous administration of 120 ml of contrast material, a thin collimation chest-CT scan was performed (120 kV, 100 mAs, collimation: 4 x 1 mm). If PE was present, or previous examinations and clinical signs suggested deep venous thrombosis (DVT), a CTP was subsequently completed. CTPs were performed using a 4 x 5 mm protocol (120 kV, 170 mAs). Venous phase scanning, starting from the pelvic crest, was completed in the popliteal fossa three minutes after contrast material injection. In 73 extremities, CTP were compared to the results of ultrasound, phlebography and autopsy. Scan ranges were documented in all patients. Cumulative doses were calculated for male and female subgroups.

RESULTS

62 patients in our series suffered from PE and in 47 of these patients deep venous thrombosis was seen additionally. Of the 99 patients without PE, 47 also received indirect CTP. CTP confirmed the suspicion and extent of DVT in 8 patients. Only in 2 of 39 patients (5.1 %) was previously unknown DVT found, despite the exclusion of PE. Regarding DVT, sensitivity was 94.3 % and specificity was 92.1 % for indirect CTP. Cumulative chest CT doses averaged 3.3 mSv for males and 4.2 mSv for females, the calculated CTP dosage was 9.3 mSv (according to ICRP 60).

CONCLUSIONS

The examination protocol presented is suitable for clinical usage in patients with suspected PE. If PE is confirmed, indirect CTP is justified, so that detailed information of the venous system can be obtained. However, the relatively high radiation dosage of an additional CTP requires a strict indication regiment in patients with a negative CTA.

Combined image interpretation of computed tomography and hybrid PET in head and neck cancer

AIM

Evaluation of potential synergistic effects of combined image interpretation of FDG PET using a gamma camera modified for coincidence detection (hybrid PET) and computed tomography (CT) and comparison of the diagnostic accuracy of hybrid PET and dedicated PET in patients with head and neck cancer.

METHODS

Forty-two patients with suspected primary or recurrent cancer were included. Twenty-four patients underwent dedicated PET in addition to attenuation-corrected hybrid PET using a one-day protocol.

RESULTS

Sensitivity, specificity and accuracy for detection of primary or recurrent head and neck cancer were 74, 73, and 74% for hybrid PET, 52, 82, and 60% for CT and 77, 82, and 79% for combined reading. With the combination of CT and hybrid PET all cases of recurrent disease were detected. The largest tumour not detected was 1.7 cm in diameter. Sensitivity, specificity and accuracy for the detection of neck sides with lymph node metastases were 69, 88, and 85% for hybrid PET, 62, 88, and 84% for CT, 69, 99, and 94% for combined image interpretation. With combined interpretation four involved neck sides were missed including two cases of microscopic metastases. Hybrid PET revealed concordant results to dedicated PET in all patients with respect to the detection of primary or recurrent tumour and in 45 of 48 neck sides (94%) with the same number of false negative findings.

CONCLUSION

The combination of functional information of hybrid PET and morphological information of CT by the simple approach of combined image interpretation improves the sensitivity for the detection of primary/recurrent head and neck cancer and increases the specificity of lymph node staging compared to CT alone. The accuracy of hybrid PET and dedicated PET was almost identical.