GI magnetic resonance angiography

Endoscopic treatment of gastric varices

Gastric varices (GV) are present in one in 5 patients with portal hypertension and variceal bleeding. GV bleeds tend to be more severe with higher mortality. High index of suspicion, early detection and proper locational diagnosis are important. An algorithmic approach to the management of GV bleeding prevents rebleeds and improves survival. Vasoactive drugs should be started with in 30 minutes (door to needle time) and early endotherapy be done. Cyanoacrylate injection in experienced hands achieves hemostasis in >90% patients. A repeat session is sometimes needed for complete obturation of GV. Transjugular intrahepatic portosystemic shunt and balloon retrograde transvenous obliteration are effective rescue options. Secondary prophylaxis of GV bleeding is done with beta-blocker and endotherapy.

Carbon dioxide-based portography: an alternative to conventional imaging with the use of iodinated contrast medium

BACKGROUND AND AIM

To clarify the efficacy of carbon dioxide (CO(2)) as a contrast material to evaluate portal vein images by percutaneous transhepatic portography (PTP).

METHODS

Twenty patients (38-76 years; male 13, female 7) with chronic liver diseases were the subjects of this prospective study. Portal venous opacification by PTP was compared between CO(2)-based images and iodinated contrast medium (ICM)-based images by two independent reviewers, according to the three-grade scoring; 0 for none, 1 for weak and 2 for sufficient.

RESULTS

Total scores of extrahepatic portal veins (137 for CO(2), 93 for ICM), collateral vessels (64 for CO(2), 60 for ICM) and intrahepatic portal veins (69 for CO(2), 76 for ICM) were not statistically significant between CO(2)-based and ICM-based images (P = 0.0623). Sufficient opacification of superior mesenteric vein was more frequent on CO(2)-based images (none 0, weak 4, sufficient 16) than ICM-based images (none 19, weak 0, sufficient 1; P < 0.0001). The score was not statistically significant between CO(2)-based and ICM-based images in portal trunk, splenic vein, inferior mesenteric vein and other collateral vessels. Although opacification grade in the intrahepatic left portal vein was not statistically significant between CO(2)-based and ICM-based images (P = 0.1515), weak opacification was significantly frequent on CO(2)-based images (weak 10, sufficient 10) compared to ICM-based images (weak 0, sufficient 20; P = 0.0003) in the intrahepatic right portal vein. Inter-reviewer agreement was excellent between the two reviewers for CO(2)-based images (kappa = 0.913) and ICM-based images (kappa = 0.924).

CONCLUSIONS

Carbon dioxide may be a first-line contrast material for evaluating portal vein images by PTP.

Hemodynamic features of gastrorenal shunt: a Doppler study in cirrhotic patients with gastric fundal varices

RATIONALE AND OBJECTIVES

Little is known about the hemodynamics of gastrorenal shunt (GRS), a major drainage route of gastric fundal varices (FV), in patients with FV. The aim of this study was to clarify the hemodynamic features of GRS on Doppler sonography in relation to the grading and bleeding of FV.

MATERIALS AND METHODS

The study subjects consisted of 69 cirrhotic patients with FV. Diameter, flow velocity (FVe), and flow volume (FVo) of GRS were measured by Doppler ultrasound (US). The detection rate was compared to contrast-enhanced computed tomography (CECT), and percutaneous transhepatic portography (PTP) was used in six patients without GRS on CECT.

RESULTS

The use of CECT detected GRS in 60 of 69 patients, and US, 58 of 69 patients. A false-negative result for detecting GRS on both CECT and US was found in one patient after PTP. The diameter, FVe, and FVo of GRS increased according to the endoscopic grade of FV: F1 (7.2+/-1.3 mm, 9.8+/-1.1 cm/s, 358.3+/-123.4 ml/min), F2 (9.9+/-3.3 mm, 12.8+/-5.1 cm/s, 701.7+/-411.3 ml/min), and F3 (11.8+/-2.4 mm, 17.9+/-8.3 cm/s, 1706.6+/-989.5 ml/min). A significant difference was seen between F1 and F3 (diameter, P=.0022; FVe, P=.0133; FVo, P=.0007) and between F2 and F3 (FVe, P=.0112; FVo, P<.0001). FVe of GRS was significantly higher in bleeders (16.7+/-8.1 cm/s) than in nonbleeders (12.2+/-5.4 cm/s, P=.017), whereas the diameter and FVo were not significant.

CONCLUSION

Hemodynamics of GRS on Doppler sonograms reflected the grading and bleeding of FV. Doppler US may be valuable as a noninvasive method to evaluate the severity of FV.

Comparison of capsule endoscopy and magnetic resonance (MR) enteroclysis in suspected small bowel disease

BACKGROUND AND AIMS

Small bowel MR enteroclysis and wireless capsule endoscopy (WCE) are new diagnostic tools for the investigation of the small bowel. The aim of this study was to compare the diagnostic yield of WCE with MR enteroclysis in the detection of small bowel pathologies.

METHODS

A total of 36 patients were included in the study. Indications for imaging of the small bowel were proven or suspected small bowel Crohn's disease (CD; n=18), obscure gastrointestinal (GI) bleeding (n=14) and tumour surveillance (n=4).

RESULTS

In patients with Crohn's disease WCE detected significantly more inflammatory lesions in the first two segments of the small bowel compared with MR enteroclysis (12 patients vs. 1 patient, p=0.016). In 5 out of 14 (36%) patients with GI bleeding, angiodysplasia was detected as a possible bleeding source. Three of these patients had active bleeding sites detected by WCE. One patient had scattered inflammation of the mucosa. MR enteroclysis did not reveal any intestinal abnormalities in this patient group. MR enteroclysis provided extraintestinal pathologies in 10 out of 36 (28%) patients.

CONCLUSION

In patients with Crohn's disease WCE revealed significantly more inflammatory lesions in the proximal and middle part of the small bowel in comparison to MR enteroclysis, whereas in patients with obscure GI bleeding WCE was superior to MR enteroclysis.

Comparison of image quality between electronic and mechanical radial scanning echoendoscopes in pancreatic diseases

BACKGROUND AND AIM

It is common knowledge that endoscopic ultrasonography (EUS) can accurately diagnose pancreatic diseases. Echoendoscopes for EUS are roughly classified into two categories, the mechanical radial scanning echoendoscope (MR-ES) and the electronic linear array echoendoscope, both of which have their merits and demerits. In 2000, a newly designed echoendoscope, the electronic radial scanning echoendoscope (ER-ES), appeared. The aim of the present study was to compare B-mode image quality between the ER-ES and the MR-ES in pancreatic diseases.

METHODS

Patients with pancreatic diseases (30 cystic diseases and 22 solid diseases) underwent EUS with both ER-ES and MR-ES. The B-mode images obtained using both echoendoscopes were graded using a scoring system and statistically analyzed. The assessed point for cystic lesions was the existence of mechanical-noise-like ring-like artifacts derived using multiple reflections ('ring-down'), grating robe and so on, and that for solid lesions was the scale of penetration. The authors compared maneuverability, endurance and endoscopic images between the two types of echoendoscopes.

RESULTS

The ER-ES had a significantly higher score than the MR-ES (P < 0.05) in the analysis of both cystic and solid diseases. There was no apparent difference as to maneuverability, endurance and endoscopic images.

CONCLUSION

Ultrasound images acquired by ER-ES appear better compared with those acquired by MR-ES.

Preclinical study of endoscopic ultrasonography with electronic radial scanning echoendoscope

BACKGROUND

To evaluate the imaging possibility of a newly designed electronic radial scanning echoendoscope (ER-ES).

METHODS

In the in vivo study of swine, we obtained B-mode endoscopic ultrasonography (EUS) images of the gastric and gallbladder (GB) walls and checked the ability to detect Doppler signals using ER-ES and electronic linear array echoendoscope (EL-ES). Furthermore, in the ex vivo study of swine, B-mode EUS images of fixed gastric and GB wall specimens were obtained using ER-ES, EL-ES and mechanical radial scanning echoendoscope (MR-ES). In the study of resected human specimens, we obtained B-mode EUS images of five resected GB specimens (three normal GB, one cholecystitis and one cancerous) using the three types of echoendoscope.

RESULTS

In the in vivo study of swine, ER-ES and EL-ES depicted the gastric walls as five-layered, and the GB walls as single-layered structures. The ability to detect Doppler signals was equal between ER-ES and EL-ES. In the ex vivo study of swine, ER-ES, MR-ES and EL-ES equally delineated the gastric walls as five-layered and GB walls as three-layered structures. In the study of resected human specimens, results demonstrated the normal GB walls as three-layered, the cholecystitis as a combination of outer high-echoic and inner low-echoic layers, and the cancer as a protruded tumor.

CONCLUSIONS

We conclude that ER-ES has faculties for making B-mode images as well as EL-ES and MR-ES. In addition, in the in vivo study, ER-ES can analyze blood flow information as well as EL-ES.