Does secretin stimulation add to magnetic resonance cholangiopancreatography in characterising pancreatic cystic lesions as side-branch intraductal papillary mucinous neoplasm?

Objectives

To assess the value of secretin during magnetic resonance cholangiopancreatography (MRCP) in demonstrating communication between cystic lesions and the pancreatic duct to help determine the diagnosis of side-branch intraductal papillary mucinous neoplasm (SB-IPMN).

Methods

This is an IRB-approved, HIPAA-compliant retrospective study of 29 SB-IPMN patients and 13 non-IPMN subjects (control) who underwent secretin-enhanced MRCP (s-MRCP). Two readers blinded to the final diagnosis reviewed three randomised image sets: (1) pre-secretin HASTE, (2) dynamic s-MRCP and (3) post-secretin HASTE. Logistic regression, generalised linear models and ROC analyses were used to compare pre- and post-secretin results.

Results

There was no significant difference in median scores for the pre-secretin [reader 1: 1; reader 2: 2 (range -2 to 2)] and post-secretin HASTE [reader 1: 1; reader 2: 1 (range -2 to 2)] in the SB-IPMN group (P?=?0.14), while the scores were lower for s-MRCP [reader 1: 0.5 (range -2 to 2); reader 2: 0 (range -1 to 2); P?=?0.016]. There was no significant difference in mean maximum diameter of SB-IPMN on pre- and post-secretin HASTE, and s-MRCP (P?>?0.05).

Conclusion

Secretin stimulation did not add to MRCP in characterising pancreatic cystic lesions as SB-IPMN.

Key Points

? Magnetic resonance cholangiopancreatography (MRCP) is used to evaluate pancreatic cystic lesions. ? Intraductal papillary mucinous neoplasm (IPMN) is a type of pancreatic cystic neoplasm. ? Secretin administration does not facilitate the diagnosis of IPMN on MRCP.     

Evaluation of the biliary and pancreatic system with 2D SSFSE, breathhold 3D FRFSE and respiratory-triggered 3D FRFSE sequences

Purpose

The authors compared biliary and pancreatic imaging obtained through 2D single-shot fast spin-echo (SSFSE), breath-hold 3D fast recovery fast spin-echo (FRFSE) and respiratory-triggered 3D FRFSE sequences.

Materials and methods

A total of 106 magnetic resonance cholangiopancreatography (MRCP) examinations performed between December 2007 and September 2008 were evaluated with a comparison of 2D SSFSE (thin section and thick slab), breath-hold 3D FRFSE and respiratory-triggered 3D FRFSE sequences. The biliary tract was divided into seven segments: right hepatic duct, left hepatic duct, common hepatic duct, cystic duct, common bile duct, cystic duct junction and biliarypancreatic confluence. The main pancreatic duct was divided into three segments (head, body and tail). Visualisation of biliary variants was also compared. Two blinded radiologists evaluated segment visibility using a quantitative scale. The Student’s t test for paired samples was used for statistical analysis.

Results

Compared with 2D SSFSE, respiratory-triggered 3D FRFSE sequences showed better visibility of the right hepatic duct (p=0.0277), the cystic duct (p=0.0081), the cystic duct junction (p=0.0010), the biliary-pancreatic confluence (p=0.0334) and biliary variants (p=0.0198). In the comparison between breath-hold 3D FRFSE and 2D SSFSE, a significant statistical difference was found in visualisation of the cystic duct (p=0.027), the cystic duct junction (p=0.020), the biliary-pancreatic confluence (p=0.0338) and biliary variants (p=0.0311).

Conclusions

Three-dimensional FRFSE offers a significant benefit over conventional 2D imaging.     

Anatomic variants of the pancreatic duct and their clinical relevance: an MR-guided study in the general population

Objectives

To investigate the frequency of pancreatic duct (PD) variants and their effect on pancreatic exocrine function in a population-based study using non-invasive secretin-stimulated magnetic resonance cholangiopancreatography (sMRCP).

Methods

Nine hundred and ninety-five volunteers, 457 women and 538 men, aged 51.9?±?13.4 years, underwent navigator-triggered, T2-weighted, 3D turbo spin echo MRCP on a 1.5 T system after 1 unit/kg secretin administration. Two readers evaluated images for PD variants. Pancreatic exocrine function and morphological signs of chronic pancreatitis such as abnormalities of the main PD, side branch dilatation, and pancreatic cysts were evaluated and related to PD variants using a Kruskal-Wallis test and post hoc analysis.

Results

Of all sMRCP, 93.2 % were of diagnostic quality. Interobserver reliability for detection of PD variants was found to be kappa 0.752 (95 %CI, 0.733 – 0.771). Normal PD variants were observed in 90.4 % (n?=?838/927). Variants of pancreas divisum was identified in 9.6 % (n?=?89/927). Abnormalities of the main PD, side branch dilatation, and pancreatic cysts were observed in 2.4 %, 16.6 %, and 27.7 %, respectively, and were not significantly different between pancreas divisum and non-divisum group (P?=?0.122; P?=?0.152; P?=?0.741). There was no association between PD variants and pancreatic exocrine function (P?=?0.367).

Conclusion

PD variants including pancreas divisum are not associated with morphological signs of chronic pancreatitis or restriction of pancreatic exocrine function.

Key Points

? MRCP allows the evaluation of pancreatic duct variants and morphological change. ? Pancreatic duct variants are not associated with morphological signs of chronic pancreatitis. ? Pancreas divisum is not accompanied by restriction of pancreatic exocrine function. ? Pancreatic duct variants including pancreas divisum are limited in their clinical relevance.     

Repeated Intra-Arterial Therapy via the Cystic Artery for Hepatocellular Carcinoma

Purpose

To evaluate the postprocedural imaging findings and safety of repeated intra-arterial therapy via the cystic artery in patients with hepatocellular carcinoma (HCC).

Methods

This retrospective study was approved by our institutional review board. From February 2002 to January 2012, we performed repeated (two or more) chemotherapeutic infusion or chemoembolization via the cystic artery using iodized oil in 132 patients with HCCs. Computed tomographic (CT) scans, digital subtraction angiograms, and medical records were retrospectively reviewed by consensus.

Results

A total of 340 sessions of intra-arterial therapy (160 sessions of chemotherapeutic infusion and 180 sessions of chemoembolization) via the cystic artery were undertaken in 132 patients. Fifty-five of 132 patients received both chemotherapeutic infusion and chemoembolization. The incidence of gallbladder wall thickening on follow-up contrast-enhanced CT was significantly higher in chemoembolization (48 of 180, 26.7 %) than in chemotherapeutic infusion (27 of 160, 16.9 %) (P = 0.035). Persistent gallbladder wall thickening was more frequently observed in chemoembolization (48 of 107, 44.9 %) than in chemotherapeutic infusion (27 of 90, 30 %) (P = 0.039). The major complication rate was 15 of 340 sessions (4.4 %) with 11 of 132 patients (8.3 %). Acute cholecystitis, which was related to intra-arterial therapy via the cystic artery, developed in two patients and was managed by conservative treatment.

Conclusion

HCC supplied by the cystic artery can be safely treated by repeated intra-arterial chemotherapeutic infusion or chemoembolization using iodized oil through the cystic artery.     

Comparative diagnostic evaluation with MR cholangiopancreatography, ultrasonography and CT in patients with pancreatobiliary disease

Purpose

The aim of this study was to directly compare the results of magnetic resonance cholangiopancreatography (MRCP) with those of ultrasonography (US) and multislice computed tomography (MSCT) in the diagnosis of pancreaticobiliary diseases.

Materials and methods

A total of 70 patients (41 men, 29 women) aged 22-89 years were studied either before (n=59) or after cholecystectomy (n=11) for biliary lithiasis. Clinical signs and symptoms were jaundice (n=15), abdominal pain (n=37) and proven biliary lithiasis (n=18). MRCP was performed in all patients, whereas abdominal US was performed in 55 (group 1) and MSCT in 37 (group 2) patients. A regional evaluation of the main structures of the pancreaticobiliary system was performed: gallbladder and cystic duct, intra- and extrahepatic bile ducts and main pancreatic duct. Histology (n=27), biopsy (n=5), endoscopic retrograde cholangiopancreatography (ERCP) (n=28) and/or clinical-imaging follow-up (n=10) were considered standards of reference. In particular, patients were classified as showing benign (n=47) or malignant (n=12) lesions or normal biliary anatomy (n=11).

Results

In group 1, the results of MRCP and US were concordant in the majority (92%) of cases; however, statistically significant discordance (p<0.01) was found in the evaluation of the extrahepatic ducts, with nine cases (16%) of middle-distal common bile duct stones being detected on MRCP only. In group 2, the results of MRCP and MSCT were also concordant in most cases (87%). However, findings were significantly discordant when the intra- and extrahepatic ducts were analysed, with seven (19%) and six (16%) cases, respectively, of lithiasis being detected on MRCP only (p<0.01 for both).

Conclusions

The results of our study confirm the diagnostic potential of MRCP in the study of the pancreaticobiliary duct system. In particular, the comparison between MRCP and US and MSCT indicates the superiority of MRCP in evaluating bile ducts and detecting stones in the common bile duct.     

MR signal changes on hepatobiliary imaging after oral ingestion of manganese chloride tetrahydrate: preliminary examination

Purpose

To clarify magnetic resonance (MR) signal changes during hepatobiliary imaging after oral ingestion of manganese chloride tetrahydrate (Bothdel Oral Solution 10) (MCT).

Materials and methods

Subjects were an MCT phantom, 10 healthy volunteers and 155 consecutive patients. The phantom study evaluated the relationship between MCT concentration and MR signal intensity. The volunteer study qualitatively and quantitatively analyzed the time course of hepatobiliary imaging immediately and 30, 60, 120 and 180 min after oral ingestion of MCT. The clinical study analyzed the incidence and factors affecting signal changes on additional MRCP with MCT as a final scan during routine clinical MRCP.

Results

In the phantom study, a significant and excellent positive linear correlation was found between MCT concentration and R2* measurement (r ² = 0.996, p < 0.01). In the volunteer study, biliary imaging and hepatic R2* values changed significantly after oral ingestion of MCT (p < 0.05). In the clinical study, a signal drop on MRCP with MCT was visually confirmed in 14.8 % of cases. Multivariate logistic regression found no factors significantly affecting signal drop.

Conclusion

We recommend performing MRCP early after oral ingestion of MCT, because signal changes occur even during routine MRCP.     

The Origin of the Cystic Artery Supplying Hepatocellular Carcinoma on Digital Subtraction Angiography in 311 Patients

Purpose

This study was designed to investigate the prevalence and patterns of origin of the cystic artery using selective angiography images obtained during chemoembolization.

Methods

Between March 2007 and January 2012, 326 patients with hepatocellular carcinoma supplied by the cystic artery were treated by chemoembolization through the cystic artery. Fifteen patients were excluded due to the difficulty in determining the origin of their cystic arteries. Thus, a total of 311 patients were included in this study. Digital subtraction angiography images were reviewed retrospectively by consensus.

Results

A total of 112 (36 %) patients had a total of 121 variant hepatic arteries. Double cystic arteries were present in 46 (14.8 %) patients, and total 357 cystic arteries were observed. The origin sites of the cystic artery were the following: the right hepatic artery (n = 248), the anterior sectional artery (n = 44), the segment IV artery (n = 21), the posterior sectional artery (n = 10), the left hepatic artery (n = 8), the proper hepatic artery (n = 8), the gastroduodenal artery (n = 4), and others (n = 14). In total, 103 (33 %) patients had at least one cystic artery arising from arteries other than the right hepatic artery. In patients with right hepatic artery variations, the cystic artery more frequently originated from arteries other than the right hepatic artery (p = 0.003).

Conclusions

The most and second most common origins of the cystic artery are the right hepatic artery and the anterior sectional artery. In addition, the origin site of the cystic arteries is influenced by variations of hepatic arteries.     

Postprandial changes in secretory flow of pancreatic juice in the main pancreatic duct: evaluation with cine-dynamic MRCP with a spatially selective inversion-recovery (IR) pulse

Purpose

To evaluate the influence of oral ingestion on the secretory flow dynamics of physiological pancreatic juice within the main pancreatic duct in healthy subjects by using cine-dynamic MRCP with spatially-selective inversion-recovery (IR) pulse non-invasively.

Materials and methods

Thirty-eight healthy subjects were investigated. MRCP with spatially-selective IR pulse was repeated every 15 s for 5 min to acquire a total of 20 images (cine-dynamic MRCP). A set of 20 MRCP images was repeatedly obtained before and after liquid oral ingestion every 7 min (including 2-min interval) for 40 min (a total of seven sets). Secretion grade of pancreatic juice on cine-dynamic MRCP was compared before and after oral ingestion using the nonparametric Wilcoxon signed-rank test.

Results

Median secretion grades of pancreatic juice at 5 min (score?=?2.15), 12 min (score?=?1.95) and 19 min (score?=?2.05) after ingestion were significantly higher than that before ingestion (score?=?1.40) (P?=?0.004, P?=?0.032, P?=?0.045, respectively). Secretion grade of pancreatic juice showed a maximum peak of 2.15 at 5 min after ingestion. Thereafter, the secretion grade of pancreatic juice tended to gradually decline.

Conclusion

Non-invasive cine-dynamic MRCP using spatially-selective IR pulse showed potential for evaluating postprandial changes in the secretory flow dynamics of pancreatic juice as a physiological reaction.

Key Points

? Secretion grade of pancreatic juice at cine-dynamic MRCP after ingestion was evaluated. ? Secretion grade was significantly increased within 19 min after liquid meal ingestion. ? Secretion grade showed maximum peak of 2.15 at 5 min after ingestion. ? Postprandial changes in pancreatic juice flow can be assessed by cine-dynamic MRCP.

     

Contrast-enhanced free-breathing 3D T1-weighted gradient-echo sequence for hepatobiliary MRI in patients with breath-holding difficulties

Objective

Evaluate the image quality and diagnostic performance of a free-breathing 3D-gradient-echo sequence with radial acquisition (rGRE) compared with a Cartesian breath-hold 3D-GRE (cGRE) sequence on hepatobiliary phase MRI in patients with breath-holding difficulties.

Methods

Twenty-eight consecutive patients (15 males; mean age 61?±?11.9 years) were analysed in this retrospective IRB-approved study. Breath-holding difficulties during gadoxetate-disodium-enhanced liver MRI manifested as breathing artefacts during dynamic-phase imaging. MRI included axial and coronal cGRE and a radially sampled rGRE sequence during the hepatobiliary phase. Two radiologists independently evaluated cGRE and rGRE images for image quality, liver lesion detection and conspicuity, and bile duct conspicuity on a four-point scale.

Results

Liver edge sharpness was significantly higher on rGRE images (P?<?0.001). Overall image quality was slightly but significantly higher for rGRE than for cGRE (P?<?0.001 and P?=?0.039). Bile duct conspicuity scores of rGRE and cGRE were not significantly different. Sensitivity for detection of the 26 liver lesions was similar for rGRE and cGRE (81-77 % and 73-77 %, P?=?0.5 and 1.0). Lesion conspicuity scores were significantly higher for rGRE for one reader (P?=?0.012).

Conclusion

In patients with breath-holding difficulties, overall image quality and liver lesion conspicuity on hepatobiliary phase MRI can be improved using the rGRE sequence.

Key Points

? Patients with diminished breath-holding capacities present a major challenge in abdominal MRI. ? A free-breathing sequence for hepatobiliary-phase MRI can improve image quality. ? Further advances are needed to reduce acquisition time of the free-breathing gradient-echo sequence.     

GRASE Revisited: breath-hold three-dimensional (3D) magnetic resonance cholangiopancreatography using a Gradient and Spin Echo (GRASE) technique at 3T

Objective

To evaluate the clinical feasibility and image quality of breath-hold (BH) three-dimensional (3D) magnetic resonance cholangiopancreatography (MRCP) using a gradient and spin-echo (GRASE) technique compared to the conventional 3D respiratory-triggered (RT)-MRCP using a turbo spin-echo (TSE) sequence at 3 T.

Methods

Sixty-six patients underwent both 3D RT-TSE-MRCP and 3D BH-GRASE-MRCP at 3 T. Three radiologists independently reviewed the visualisation of biliary and pancreatic ducts, image blurring, and overall image quality of the two data sets using four- or five-point scales. The numbers of scans with non-diagnostic or poor image quality were compared between the two scans.

Results

The 3D BH-GRASE-MRCP had a significantly better image quality (3.69 ± 0.77 vs. 3.30 ± 1.18, p = 0.005) and less image blurring (3.23 ± 0.94 vs. 3.65 ± 0.57, p = 0.0003) than the 3D RT-TSE-MRCP. In detail, 3D BH-GRASE-MRCP better depicted the common bile duct, cystic duct, and bilateral first intrahepatic duct (all ps < 0.05). The number of scans with non-diagnostic or poor image quality significantly decreased with 3D BH-GRASE-MRCP compared with 3D RT-TSE-MRCP [19.7% (13/66) vs. 1.5% (1/66), p = 0.002].

Conclusion

The 3D BH-GRASE-MRCP provided better image quality and a reduced number of non-diagnostic images compared to 3D RT-TSE-MRCP.

Key points

? The GRASE technique enabled 3D MRCP acquisition within a single breath-hold. ? The short acquisition time of 3D BH-GRASE-MRCP significantly reduced image blurring. ? The 3D BH-GRASE-MRCP had a better image quality than 3D RT-TSE-MRCP. ? The number of non-diagnostic scans was reduced with 3D BH-GRASE-MRCP.

     

Selective Internal Radiotherapy (SIRT) of Hepatic Tumors: How to Deal with the Cystic Artery

Purpose

Selective internal radiotherapy (SIRT) with the beta emitter yttrium-90 (Y90) is a rapidly developing therapy option for unresectable liver malignancies. Nontarget irradiation of the gallbladder is a complication of SIRT. Thus, we aimed to assess different strategies to avoid infusion of Y90 into the cystic artery (CA).

Methods

After hepatic digital subtraction angiography and administration of technetium-99m-labeled human serum albumin (⁹⁹mTc-HSA), 295 patients with primary or secondary liver tumors underwent single-photon emission computed tomography/computed tomography (SPECT/CT). Different measures were taken before repeated Y90 mapping and SIRT to avoid unintended influx into the CA where necessary. Clinical symptoms, including pain, fever, or a positive Murphy sign, were assessed during patient follow-up.

Results

A significant ⁹⁹mTc-HSA accumulation in the gallbladder wall (higher ⁹⁹mTc-HSA uptake than in normal liver tissue) was seen in 20 patients. The following measures were taken to avoid unintended influx into the CA: temporary/permanent occlusion of the CA with gelfoam (n = 5)/microcoil (n = 1), induction of vasospasm with a microwire (n = 4), or altering catheter position (n = 10). Clinical signs of cholecystitis were observed in only one patient after temporary CA occlusion with gelfoam and were successfully treated by antibiotics. Cholecystectomy was not required for any patient.

Conclusion

It is important to identify possible nontarget irradiation of the gallbladder. The risk for radiation-induced cholecystitis can be easily minimized by temporary or permanent CA embolization, vasospasm induction, or altering the catheter position.     

The efficiency of apparent diffusion coefficient quantification in diagnosis of acute cholecystitis and in differentiation of cholecystitis from extrinsic benign gallbladder wall thickening

Purpose

The aim of the current study was to assess the efficiency of the apparent diffusion coefficient (ADC) measurement in diagnosis of acute cholecystitis and in differentiation of cholecystitis from extrinsic benign gallbladder wall thickening.

Methods

Forty patients who were diagnosed to have acute cholecystitis by ultrasonographic examination were included in this study. The control group consisted of 18 patients without symptoms of gallstones and cholecystitis whose gallbladder walls were thickened due to cirrhotic ascites. Both groups were examined using diffusion weighted imaging, and the mean ADC values were compared using Student’s t-test.

Results

The diagnoses of the 40 patients were proven by histopathological examination. The mean ADC values of patients diagnosed with cholecystitis (1.68 ± 0.36 × 10^?3 mm²/s) were significantly lower than the mean ADC values of the control group (2.35 ± 0.24 × 10^?3 mm²/s) (p < 0.05). Receiver operating characteristics curve analysis based on ADC revealed a cut-off value of 2.04 × 10^?3 mm²/s for the diagnosis of cholecystitis, with a sensitivity of 94 % and a specificity of 89.7 %.

Conclusion

ADC value quantification may be an efficient method for making a diagnosis of cholecystitis and in differential diagnosis of cholecystitis from the extrinsic benign gallbladder wall thickening that can be seen during the course of cirrhotic ascites.     

Percutaneous Transhepatic Removal of Bile Duct Stones: Results of 261 Patients

Purpose

To determine the effectiveness of percutaneous transhepatic removal of bile duct stones when the procedure of endoscopic therapy fails for reasons of anatomical anomalies or is rejected by the patient.

Methods

Between April 2001 and May 2010, 261 patients (138 male patients and 123 female patients; age range, 14–92 years; mean age, 64.6 years) with bile duct stones (common bile duct [CBD] stones = 248 patients and hepatolithiasis = 13 patients) were included in the study. First, percutaneous transhepatic cholangiography was performed and stones were identified. Percutaneous transhepatic balloon dilation of the papilla of Vater was performed. Then stones were pushed out into the duodenum with a Fogarty balloon catheter. If the stone diameter was larger than 15 mm, then basket lithotripsy was performed before balloon dilation.

Results

Overall success rate was 95.7%. The procedure was successful in 97.5% of patients with CBD stones and in 61.5% of patients with hepatolithiasis. A total of 18 (6.8%) major complications, including cholangitis (n = 7), subcapsular biloma (n = 4), subcapsular hematoma (n = 1), subcapsular abscess (n = 1), bile peritonitis (n = 1), duodenal perforation (n = 1), CBD perforation (n = 1), gastroduodenal artery pseudoaneurysm (n = 1), and right hepatic artery transection (n = 1), were seen after the procedure. There was no mortality.

Conclusion

Our experience suggests that percutaneous transhepatic stone expulsion into the duodenum through the papilla is an effective and safe approach in the nonoperative management of the bile duct stones. It is a feasible alternative to surgery when endoscopic extraction fails or is rejected by the patient.     

Optimising diffusion-weighted MR imaging for demonstrating pancreatic cancer: a comparison of respiratory-triggered, free-breathing and breath-hold techniques

Objectives

To compare respiratory-triggered, free-breathing, and breath-hold DWI techniques regarding (1) image quality, and (2) signal intensity (SI) and ADC measurements in pancreatic ductal adenocarcinoma (PDAC).

Methods

Fifteen patients with histopathologically proven PDAC underwent DWI prospectively at 1.5?T (b?=?0, 50, 300, 600 and 1,000?s/mm²) with the three techniques. Two radiologists, independently and blindly, assigned total image quality scores [sum of rating diffusion images (lesion detection, anatomy, presence of artefacts) and ADC maps (lesion characterisation, overall image quality)] per technique and ranked them. The lesion SI, signal-to-noise ratio, mean ADC and coefficient of variation (CV) were compared.

Results

Total image quality scores for respiratory-triggered, free-breathing and breath-hold techniques were 17.9, 16.5 and 17.1 respectively (respiratory-triggered was significantly higher than free-breathing but not breath-hold). The respiratory-triggered technique had a significantly higher ranking. Lesion SI on all b-values and signal-to-noise ratio on b300 and b600 were significantly higher for the respiratory-triggered technique. For respiratory-triggered, free-breathing and breath-hold techniques the mean ADCs were 1.201, 1.132 and 1.253?×?10^-3?mm²/s, and mean CVs were 8.9, 10.8 and 14.1?% respectively (respiratory-triggered and free-breathing techniques had a significantly lower mean CV than the breath-hold technique).

Conclusion

In both analyses, respiratory-triggered DWI showed superiority and seems the optimal DWI technique for demonstrating PDAC.

Key Points

? Diffusion-weighted magnetic resonance imaging is increasingly used to detect pancreatic cancer ? Images are acquired using various breathing techniques and multiple b-values ? Breathing techniques used: respiratory-triggering, free-breathing and breath-hold ? Respiratory-triggering seems the optimal breathing technique for demonstrating pancreatic cancer     

Diffusion-weighted MR imaging for the differentiation of malignant from benign strictures in the periampullary region

Objectives

To investigate the value of DWI for differentiating malignant from benign strictures in the periampullary region.

Methods

We retrospectively analysed data from 78 patients who had undergone magnetic resonance cholangiopanreatography (MRCP) and diffusion-weighted imaging (DWI), in whom biliary strictures in the periampullary region were suspected. Twenty-two malignant and 56 benign lesions were included. One radiologist compared the signal intensity of malignant and benign periampullary lesions on DWI using b?=?500 and 800 s/mm². The signal intensity of bile was also compared, and an optimal b value was determined for periampullary lesions. Two other radiologists reviewed MRCP alone and combined DWI and MRCP for the possibility of malignant periampullary lesions. Diagnostic accuracy was calculated for each reviewer by receiver operating characteristic (ROC) curve analysis.

Results

Malignant periampullary lesions more frequently appeared hyperintense than benign lesions on DWI using the two b values (P?<?0.001). Bile more frequently appeared hyperintense on DWI using b?=?500 s/mm² (87.2 %) than b?=?800 s/mm² (24.4 %). Therefore, b?=?800 s/mm² was determined as the preferred sequence. Diagnostic accuracy for malignant periampullary lesions improved for both reviewers after adding DWI; from 0.714 to 0.924 (P?=?0.006, for reviewer 1) and from 0.714 to 0.919 (P?=?0.007, reviewer 2).

Conclusions

Combined DWI with MRCP can improve the diagnostic accuracy for differentiating malignant from benign strictures in the periampullary region.

Key Points

? Diffusion-weighted magnetic resonance imaging provides yet more information about hepatobiliary structures. ? Diffusion-weighted imaging (DWI) has now been applied to the biliary tree. ? Most periampullary carcinomas appear hyperintense on high b value DWI. ? DWI can help differentiate between malignant and benign periampullary lesions.     

Safety of Chemotherapeutic Infusion or Chemoembolization for Hepatocellular Carcinoma Supplied Exclusively by the Cystic Artery

Purpose

This study was designed to evaluate the safety of chemotherapeutic infusion or chemoembolization by way of the cystic artery in patients with hepatocellular carcinoma (HCC) supplied exclusively by the cystic artery.

Methods

Between Jan 2002 and Dec 2011, we performed chemotherapeutic infusion or chemoembolization using iodized oil for the treatment of 27 patients with HCC supplied exclusively by the cystic artery. Computed tomography (CT) scans, digital subtraction angiograms, and medical records were retrospectively reviewed by consensus.

Results

The cystic artery originated from the main right hepatic artery in 24 (89 %) patients, from the right anterior hepatic artery in 2 (7 %) patients, and from the left hepatic artery in 1 (4 %) patient. Selective catheterization of the cystic artery was achieved in all patients. Superselection of tumor-feeding vessels from the cystic artery was achieved in 7 patients (26 %). Chemotherapeutic infusion was performed in 18 patients (67 %), and chemoembolization was performed in 9 patients (33 %). There were no major complications and only 2 minor complications, including vasovagal syncope and nausea with vomiting. Individual tumor response supplied exclusively by the cystic artery at the follow-up enhanced CT scan were complete response (n = 16), partial response (n = 3), and stable disease (n = 8).

Conclusion

HCC supplied exclusively by the cystic artery can be safely treated without severe complications by chemotherapeutic infusion or chemoembolization using iodized oil through the cystic artery.     

Preoperative differentiation between T1a and ≥T1b gallbladder cancer: combined interpretation of high-resolution ultrasound and multidetector-row computed tomography

Objectives

To determine the diagnostic value of combined interpretation of high-resolution ultrasound (HRUS) and multidetector-row computed tomography (MDCT) for preoperative differentiation between T1a and ≥T1b gallbladder (GB) cancer.

Methods

Eighty-seven patients with pathologically confirmed GB cancers (T1a, n?=?15; ≥T1b, n?=?72), who preoperatively underwent both HRUS and MDCT, were included in this retrospective study. Two reviewers independently determined the T-stages of the GB cancers on HRUS and MDCT using a five-point confidence scale (5, definitely T1a; 1, definitely ≥T1b). For individual modality interpretation, the lesions with scores ≥4 were classified as T1a, and, for combined modality interpretation, the lesions with all scores ≥4 in both modalities were classified as T1a. The McNemar test was used to compare diagnostic performance.

Results

The diagnostic accuracy of differentiation between T1a and ≥T1b GB cancer was higher using combined interpretation (90.8 % and 88.5 % for reviewers 1 and 2, respectively) than individual interpretation of HRUS (82.8 % and 83.9 %) or MDCT (74.7 % and 82.8 %) (P?<?0.05, reviewer 1). Combined interpretations demonstrated 100 % specificity for both reviewers, which was significantly higher than individual interpretations (P?<?0.05, both reviewers).

Conclusions

Combined HRUS and MDCT interpretation may improve the diagnostic accuracy and specificity for differentiating between T1a and ≥T1b GB cancers.

Key Points

? Differentiating between T1a and ≥T1b gallbladder cancer can help surgical planning. ? HRUS and MDCT are useful for local staging of gallbladder cancer. ? HRUS and MDCT may be synergistic for T-staging of gallbladder cancer.     

Non-hyperfunctioning neuroendocrine tumours of the pancreas: MR imaging appearance and correlation with their biological behaviour

Objective

To describe MR imaging features of non-hyperfunctioning neuroendocrine pancreatic tumours by comparing them to histopathology and to determine the accuracy of MR imaging in predicting biological behaviour.

Materials and Methods

After institutional review board approval, we retrospectively reviewed 45 patients with pathologically proven NF-NET of the pancreas and ≥1 preoperative MR/MRCP examinations. Of the NF-NETS, 29/45 (64.4 %) were G1 and 16/45 (35.5 %) were G2. Image analysis included the lesion maximum diameter, vascular encasement, extrapancreatic spread, signal intensity on T1- and T2-weighted, contrast enhancement features, and presence of metastases. Tumour vessel density was calculated on the histological specimen using a grid.

Results

The median maximum diameter of NF-NETs was 20 mm (range 5–200 mm). Eighty per cent of the NF-NETs were hypointense on T1-weighted images, 82.2 % were hyperintense on T2-weighted images, and 75.6 % were hypervascular. Overall MRI accuracy showed a mean AUC of 0.86 compared to pathology. Lesions with a maximum diameter of 30 mm irregular margins, absence of a cleavage plane with the main pancreatic duct, vascular encasement, extrapancreatic spread and abdominal metastases were significantly associated with malignant NF-NETs. No correlation was found between the tumour vessel density and contrast-enhanced MR imaging pattern.

Conclusions

Hyperintensity on T2-weighted images and iso-/hypervascularity occurred in 27/45 (60.0 %) of NF-NETs. MRI identifies malignant NF-NETs with a sensitivity of 93.3 % and a specificity of 76.9 % (AUC?=?0.85).

Key Points

? Non-hyperfunctioning neuroendocrine pancreatic tumours (NF-NET) pose a difficult diagnostic challenge. ? On T2-weighted MRI, 82.2 % of neuroendocrine tumours appeared hyperintense. ? MR imaging showed 0.94 sensitivity and 0.77 specificity in predicting biological behaviour. ? The hyper-/isointensity during dynamic MRI did not correlate with vessel density at pathology.     

Imaging findings of parapharyngeal space pleomorphic adenoma in comparison with parotid gland pleomorphic adenoma

Purpose

The purpose of this study was to assess imaging findings of parapharyngeal space (PPS) pleomorphic adenoma (PA) in comparison with parotid gland (PG) PA retrospectively.

Materials and methods

Ten patients with PPS PAs and 66 with PG PAs were included. CT and MR images were reviewed for the size, presence of calcification or cystic degeneration, and MR signal intensity of solid components.

Results

The maximum diameter was greater in PPS PA (57.3 ± 11.4 mm) than in PG PA (28.7 ± 20.8 mm) (p < 0.01). The frequency of calcification was higher in PPS PA (50 %) than in PG PA (15 %) (p < 0.05). The frequency of cystic degeneration was marginally higher in PPS PA (60 %) than in PG PA (29 %) (p = 0.058). On T2-weighted images, solid components of PPS and PG PAs showed hyperintensity in 40 and 54 %, isointensity in 50 and 32 %, and hypointensity in 10 and 14 %, respectively.

Conclusion

The size and the frequency of calcification or cystic degeneration were higher in PPS PAs than in PG PAs. Radiologists should recognize the difference in imaging features between PPS and PG PAs.     

Prostate cancer recurrence after radical prostatectomy: the role of 3-T diffusion imaging in multi-parametric magnetic resonance imaging

Objectives

To validate the role of 3-T diffusion-weighted imaging (DWI) in the detection of local prostate cancer recurrence after radical prostatectomy (RP).

Methods

T2-weighted imaging, DWI and dynamic contrast-enhanced MRI (DCE-MRI) were performed with a 3-T magnet in 262 patients after RP. Twenty out of 262 patients evaluated were excluded. MRI results were validated by prostate-specific antigen (PSA) reduction after external beam radiotherapy in group A (126 patients, local recurrence size range 4–8 mm) and by transrectal ultrasound biopsy in group B (116 patients, local recurrence size range 9–15 mm).

Results

In group A combined T2-weighted and DCE-MRI (T2+DCE) shows 98 % sensitivity, 94 % specificity and 93 % accuracy in identifying local recurrence; combined T2-weighted and DWI with a b value of 3,000 s/mm² (T2+DW3) displays 97 % sensitivity, 95 % specificity and 92 % accuracy, while with a b value of 1,000 s/mm² (T2+DW1) affords 93 % sensitivity, 89 % specificity and 88 % accuracy. In group B T2+DCE shows 100 % sensitivity, 97 % specificity and 91 % accuracy in detecting local cancer recurrence; T2+DW3 displays 98 % sensitivity, 96 % specificity and 89 % accuracy; T2+DW1 has 94 % sensitivity, 92 % specificity and 86 % accuracy.

Conclusion

DCE-MRI is the most reliable technique in detecting local prostate cancer recurrence after RP, though DWI can be proposed as a reliable alternative.

Key Points

? Diffusion-weighted magnetic resonance imaging (DWI-MRI) is being increasingly used in oncology. ? PSA analysis does not distinguish prostate cancer recurrence from distant metastasis. ? DWI-MR can diagnose local prostate cancer recurrence after radical prostatectomy. ? DWI-MR is almost comparable to DCE-MRI in detecting local recurrence.