Andreas Abildgaard

Recurrent primary sclerosing cholangitis after liver transplantation: A magnetic resonance cholangiography study with analyses of predictive factors

Primary sclerosing cholangitis (PSC) is a well‐established indication for orthotopic liver transplantation (OLT), but post‐OLT bile duct strictures complicate the outcome for these patients. These strictures might represent recurrent PSC (rPSC). To estimate the risk factors for post‐OLT non‐anastomotic bile duct strictures in PSC patients and to find their possible etiology, we performed magnetic resonance cholangiography (MRC) and angiography (MRA) in all PSC patients who had undergone OLT and were alive (median follow‐up 6.4 years, range 1.4‐15.2 years). This group of PSC patients was compared to a group of 45 non‐PSC patients who had also undergone OLT. A logistic regression analysis was performed to find predictors of rPSC. Bile duct strictures were found in 19/49 PSC patients and in 4/45 non‐PSC patients (P = 0.001). In the PSC group nine patients without other possible explanations for bile duct strictures than rPSC were identified, i.e., the estimated risk of rPSC was 9/49 (18%); surprisingly similar changes were also seen in one patient without a pre‐transplant PSC diagnosis. Severe liver disease due to rPSC was seen in 4/9 patients (one patient died and three are being evaluated for re‐OLT). Steroid‐resistant rejection was the only significant predictor for rPSC. In conclusion, our study shows that by the use of MRC we found more bile duct strictures in PSC patients post‐OLT compared to controls and that steroid‐resistant rejections was a predictor of such changes. (Liver Transpl 2005.)

Laparoscopic Versus Open Resection for Colorectal Liver Metastases: The OSLO-COMET Randomized Controlled Trial.

Objective: To perform the first randomized controlled trial to compare laparoscopic and open liver resection. Summary Background Data: Laparoscopic liver resection is increasingly used for the surgical treatment of liver tumors. However, high-level evidence to conclude that laparoscopic liver resection is superior to open liver resection is lacking. Methods: Explanatory, assessor-blinded, single center, randomized superiority trial recruiting patients from Oslo University Hospital, Oslo, Norway from February 2012 to January 2016. A total of 280 patients with resectable liver metastases from colorectal cancer were randomly assigned to undergo laparoscopic (n = 133) or open (n = 147) parenchyma-sparing liver resection. The primary outcome was postoperative complications within 30 days (Accordion grade 2 or higher). Secondary outcomes included cost-effectiveness, postoperative hospital stay, blood loss, operation time, and resection margins. Results: The postoperative complication rate was 19% in the laparoscopic-surgery group and 31% in the open-surgery group (12 percentage points difference [95% confidence interval 1.67–21.8; P = 0.021]). The postoperative hospital stay was shorter for laparoscopic surgery (53 vs 96 hours, P < 0.001), whereas there were no differences in blood loss, operation time, and resection margins. Mortality at 90 days did not differ significantly from the laparoscopic group (0 patients) to the open group (1 patient). In a 4-month perspective, the costs were equal, whereas patients in the laparoscopic-surgery group gained 0.011 quality-adjusted life years compared to patients in the open-surgery group (P = 0.001). Conclusions: In patients undergoing parenchyma-sparing liver resection for colorectal metastases, laparoscopic surgery was associated with significantly less postoperative complications compared to open surgery. Laparoscopic resection was cost-effective compared to open resection with a 67% probability. The rate of free resection margins was the same in both groups. Our results support the continued implementation of laparoscopic liver resection.

Recurrent sclerosing cholangitis or ischemic bile duct lesions—A diagnostic challenge?

A 47-year-old male diagnosed with ulcerative colitis in 1985 and primary sclerosing cholangitis in 1990 was referred for orthotopic liver transplantation in 2001. He was accepted and underwent transplantation in June 2001. He experienced one acute, steroid-sensitive rejection during the first month. A hepatic artery stenosis (Fig. 1) diagnosed 30 days following transplantation was treated twice with percutaneous transluminal angioplasty and subsequently with insertion of a stent due to restenosis. The arterial blood supply was normal thereafter. A magnetic resonance cholangiography performed in December 2001 showed no strictures in the bile tree. During the next 6 to 8 months the patient had repetitive episodes with fever, increased bilirubin, pale stools, and dark urine. His symptoms responded to antibiotic treatment. During this period he had fluctuating aminotransferases, which responded to increased orally administered corticosteroids. Since August 2002, magnetic resonance cholangiography has shown multiple biliary strictures (Fig. 2). In March 2004 a liver biopsy (Fig. 3) was highly suggestive of recurrent disease. The patient suffers from recurrent, bacterial cholangitis and is currently being evaluated for retransplantation. Both his magnetic resonance cholangiography and biopsy are strongly suggestive of recurrent sclerosing cholangitis, but according to the Mayo clinic criteria1 the arterial problem excludes him from being classified as such. This emphasizes the problem of diagnosing recurrent sclerosing cholangitis, and we believe it contributes to an underestimation of the true risk of disease recurrence. The distinction between recurrent disease and ischemic bile duct lesions is of importance when evaluating the patient for a retransplantation and possibly also for the selection of immunosupression following retransplantation.

Effect of ultrasound contrast medium in color Doppler and power Doppler visualization of blood flow in canine kidneys

Purpose: to examine the effect of an ultrasound contrast medium (UCM) in the visualization of parenchymal blood flow by means of color Doppler and power Doppler sonography. Material and Methods: Nonenhanced and UCM-enhanced Doppler images of canine kidneys were obtained in a transversal plane during various states of flow reduction in the anterior branch of the renal artery. the UCM consisted of air-filled shell-stabilized microballoons (half-life approximately 2 min). the images were evaluated blindly by 4 observers who rated the amount of flow signal in the cortex and medulla, and categorized the flow state (normal, reduced or no flow) in the anterior part of the kidney. Results: the UCM increased the area of Doppler signals in the cortex and outer medulla during normal or reduced blood flow. the border between nonperfused and normally perfused parenchyma was more distinct with the UCM. the categorization of the regional flow state was more correct with the UCM. Improvement with the UCM was greatest when the nonenhanced Doppler images had suboptimal intensity, but positive effects with the UCM were also seen in recordings with adequate precontrast intensity. Color blooming artifacts sometimes occurred on the side of the kidney facing away from the transducer. Conclusion: the UCM improved the color Doppler and power Doppler visualization of the parenchymal blood flow in the canine kidney, and allowed a more correct categorization to be made of the regional blood flow state.

Use of sonicated albumin (infoson) to enhance arterial spectral and color Doppler imaging

PurposeTo examine the effect of an ultrasound contrast medium (UCM), Infoson, on Doppler examination of stenotic arteries.MethodsStenoses were created in the common carotid artery of six piglets, and examined with spectral Doppler and color Doppler imaging during UCM infusion in the left ventricle.ResultsUCM caused a mean increase in recorded maximal systolic and end-diastolic velocities of 5% and 6%, respectively, while blood flow remained constant. Increased spectral intensity with UCM was accompanied by spectral broadening. Reduction of spectral intensity by adjustment of Doppler gain counteracted the velocity effects and the spectral broadening. With color Doppler, UCM caused dose-dependent color artifacts outside the artery. Flow in narrow stenoses could be visualized with UCM.ConclusionThe effects of UCM on velocity measurements were slight, and were related to changes in spectral intensity. With color Doppler, UCM may facilitate flow detection, but color artifacts may interfere.

Evaluation of a pressure-recording guidewire in patients with coronary arterial disease

The accuracy and feasibility of coronary arterial pressure measurements with a 0.018-in. pressure-recording guidewire (PRGW) was evaluated in patients. Transstenotic pressure gradients were measured with the PRGW and a guiding catheter, at baseline and during coronary vasodilatation. Proximal intracoronary pressure was measured with both systems before and after gradient measurements. Zero pressure was measured with the PRGW before and after intracoronary use. The average of all proximal intracoronary PRGW readings were close to guiding catheter values, but there were substantial individual deviations. Average change in proximal deviation before and after gradient measurements was -1 mm Hg, standard deviation (S.D.) 7.6, range -16 to 15. Errors in zero pressure measurements after intracoronary use (average 2.8 mm Hg, S.D. 8.8, range -9 to 35) were much greater than before use (average 0.1 mm Hg, S.D. 1.4, range -4 to 3, P < 0.001). The PRGW was successfully introduced through an 8F guiding catheter and positioned across the stenosis in 21 of 26 attempts (81%). Intracoronary advancement of the PRGW through a double-lumen multifunctional probing catheter was successful in all nine attempts. In conclusion, errors in PRGW-measurements caused uncertainty in gradient interpretation. However, we found the wire useful in several cases, particularly for exclusion of hemodynamically significant lesions. The steerability of the wire is inferior to ordinary guidewires, but it can be advanced to a distal intracoronary position through an over-the-wire catheter.

Enlargement of the thymus following chemotherapy for non-seminomatous testicular cancer

The thickness of the largest thymic lobe at computed tomography (CT) was measured retrospectively in 21 relapse free patients who had undergone chemotherapy for non-seminomatous testicular cancer. CT was performed at initial staging, at completion of chemotherapy and 3 to 12 months later. Enlargement of the thymus occurred in 7 patients, one of whom had a reduced thymic size at the first two examinations after chemotherapy. A temporary reduction was detected in another patient in whom the original thymic size was regained 19 months after start of chemotherapy. The mean age of the group with thymic enlargement was 21 years compared with a mean age of 28 years in the group with no increased size. The age difference between the two groups was significant (p less than 0.05, Mann-Whitney U test). To reduce the number of explorative thoracotomies the frequent occurrence of thymic enlargement after chemotherapy for non-seminomatous testicular cancer should be kept in mind.

Use of non-invasive imaging to monitor response to aflibercept treatment in murine models of colorectal cancer liver metastases.

The liver is the most frequent metastatic site in colorectal cancer (CRC), and relevant orthotopic in vivo models are needed to study the efficacy of anticancer drugs in the metastatic setting. A challenge when utilizing such models is monitoring tumor growth during the experiments. In this study, experimental liver metastases were established in nude mice by splenic injection of the CRC cell lines HT29 and HCT116, and the mice were treated with the antiangiogenic drug aflibercept. Tumor growth was monitored using magnetic resonance imaging (MRI) and bioluminescence imaging (BLI). Aflibercept treatment was well tolerated and resulted in increased animal survival in HCT116, but not in HT29, while inhibited tumor growth was observed in both models. Treatment efficacy was monitored with high precision using MRI, while BLI detected small-volume disease with high sensitivity, but was less accurate in end-stage disease. Apparent diffusion coefficient (ADC) values obtained by diffusion weighted MRI (DW-MRI) were highly predictive of treatment response, with increased ADC corresponding well with areas of necrosis observed by histological evaluation of aflibercept-treated xenografts. The results showed that the efficacy of the antiangiogenic drug aflibercept varied between the two models, possibly reflecting unique growth patterns in the liver that may be representative of human disease. Non-invasive imaging, especially MRI and DW-MRI, can be used to effectively monitor tumor growth and treatment response in orthotopic liver metastasis models.

Comparison of 12-Bit and 8-Bit Gray Scale Resolution in Mr Imaging of the CNS An ROC Analysis

A reduction in gray scale resolution of digital images from 12 to 8 bits per pixel usually means halving the storage space needed for the images. Theoretically, important diagnostic information may be lost in the process. We compared the sensitivity and specificity achieved by 4 radiologists in reading laser-printed films of original 12-bit MR images and cathode ray tube displays of the same images which had been compressed to 8 bits per pixel using a specially developed computer program. Receiver operating characteristic (ROC) curves showed no significant differences between film reading and screen reading. A paired 2-tailed t-test, applied on the data for actually positive cases, showed that the combined, average performance of the reviewers was significantly better at screen reading than at film reading. No such differences were found for actually negative cases. Some individual differences were found, but it is concluded that gray scale resolution of MR images may be reduced from 12 to 8 bits per pixel without any significant reduction in diagnostic information.

Management of Injury to the Common Bile Duct in a Patient with Roux-en-Y Gastric Bypass

Introduction. Most surgeons prefer Roux-en-Y hepaticojejunostomy (RYHJ) for biliary reconstruction following a common bile duct (CBD) injury. However, in patients with a Roux-en-Y gastric bypass (RYGB) a RYHJ may be technically challenging and can interfere with bowel physiology induced by RYGB. The use of a hepaticoduodenostomy (HD) resolves both these issues. Presentation of Case. We present a case of CBD injury during laparoscopic cholecystectomy one year after laparoscopic RYGB for morbid obesity. Due to adhesions and previous surgery with RYGB, we did not want to interfere with the RYGB physiology by anastomosing the CBD to the jejunum or ileum. Succeeding a full Kochers maneuver we performed biliary reconstruction by a tension-free end-to-side HD. The postoperative recovery was uneventful and the patient was discharged after eight days. At four-month follow-up, the patient had stable weight and normal laboratory test results. MRCP demonstrated normal intra- and extrahepatic bile ducts with status after HD. Discussion. We propose that HD should be considered in treatment of CBD injury in post-RYGB patients as it may reduce the risk of interfering with the post-RYGB physiology.