Sarcoidosis can cause end-stage chronic liver disease, which is often unrecognized until examination of the explanted liver. Response to conventional immunosuppression is variable and unpredictable. Transplantation is feasible and safe in this population but recurrence is possible.

BACKGROUND: The role of chemotherapy for neuroendocrine tumours remains controversial and there is no standard regimen. METHODS: We report the outcome for a consecutive series of chemonaive patients with metastatic or locally advanced neuroendocrine tumours treated with a combination of 5-fluorouracil (500 mg m À2 ), cisplatin (70 mg m À2 ) and streptozocin (1000 mg m À2 ) (FCiSt) administered three weekly for up to six cycles. Patients were assessed for radiological response, toxicity and survival. RESULTS: In the 79 patients assessable for response, treatment with FCiSt was associated with an overall response rate of 33% (38% for pancreatic primary sites and 25% for non-pancreatic primary sites). Stable disease occurred in a further 51%, with progression in 16%. The median time to progression was 9.1 months and median overall survival was 31.5 months. The most common grade 3 -4 toxicity was neutropaenia (28% patients) but grade 3 -4 infection was rare (7%). The most frequent non-haematological grade 3 -4 toxicity was nausea and vomiting (17%). Prognostic factors included Ki-67, mitotic index, grade and chromogranin A, whereas response to chemotherapy was predicted by mitotic index, grade and a-fetoprotein. CONCLUSION: FCiSt is an effective regimen for neuroendocrine tumours with an acceptable toxicity profile. Grade and mitotic index are the best predictors of response.

Objective: Primary gastrointestinal neuropathies are a heterogeneous group of enteric nervous system (ENS) disorders that continue to cause difficulties in diagnosis and histological interpretation. Recently, an international working group published guidelines for histological techniques and reporting, along with a classification of gastrointestinal neuromuscular pathology. The aim of this article was to review and summarize the key issues for pediatric gastroenterologists on the diagnostic workup of congenital ENS disorders. In addition, we provide further commentary on the continuing controversies in the field. Results: Although the diagnostic criteria for Hirschsprung disease are well established, those for other forms of dysganglionosis remain ill-defined. Appropriate tissue sampling, handling, and expert interpretation are crucial to maximize diagnostic accuracy and reduce interobserver variability. The absence of validated age-related normal values for neuronal density, along with the lack of correlation between clinical and histological findings, result in significant diagnostic uncertainties while diagnosing quantitative aberrations such as hypoganglionosis or ultrashort Hirschsprung disease. Intestinal neuronal dysplasia remains a histological description of unclear significance. Conclusions:The evaluation of cellular quantitative or qualitative abnormalities of the ENS for clinical diagnosis remains complex. Such analysis should be carried out in laboratories that have the necessary expertise and access to their own validated reference values.

Portal vein embolisation (PVE) is an effective method of increasing future liver remnant (FLR) but may stimulate tumour growth. The effect of periprocedure chemotherapy has not been established. 15 consecutive patients underwent PVE prior to hepatic resection for colorectal liver metastases with a FLR <30% of tumour-free liver (TFL). Liver and tumour volumes pre-PVE and 6 weeks post-PVE were calculated by CT or MRI volumetry and correlated with the periprocedure chemotherapy regimen. PVE increased the FLR from 18+/-5% of TFL to 27+/-8% post-PVE (p<0.01). Post-PVE chemotherapy did not prevent hypertrophy of the FLR but the volume increase with chemotherapy (median 89 ml, range 7-149 ml) was significantly reduced (median 135 ml, range 110-254 ml without chemotherapy) (p = 0.016). Tumour volume (TV) decreased in those receiving post-PVE chemotherapy (median TV decrease 8 ml, range -77 ml to +450 ml) and increased without chemotherapy (median TV increase 39 ml, range -58 ml to +239 ml). Of the 15 patients, eight underwent resection; four were not resected due to disease progression and three due to insufficient hypertrophy of the FLR. PVE increased the FLR by an average of 9% allowing resection in 50% of patients. Periprocedure chemotherapy did not prevent but did reduce hypertrophy. A trend towards tumour regression was observed.