CoIn: Correlation Induced Clustering for Cognition of High Dimensional Bioinformatics Data.

Analysis of high dimensional biomedical data such as microarray gene expression data and mass spectrometry images, is crucial to provide better medical services including cancer subtyping, protein homology detection, etc. Clustering is a fundamental cognitive task which aims to group unlabeled data into multiple clusters based on their intrinsic similarities. However, for most clustering methods, including the most widely used K-means algorithm, all features of the high dimensional data are considered equally in relevance, which distorts the performance when clustering high-dimensional data where there exist many redundant variables and correlated variables. In this paper, we aim at addressing the problem of the high dimensional bioinformatics data clustering and propose a new correlation induced clustering, CoIn, to capture complex correlations among high dimensional data and guarantee the correlation consistency within each cluster. We evaluate the proposed method on a high dimensional mass spectrometry dataset of liver cancer tumor to explore the metabolic differences on tissues and discover the intra-tumor heterogeneity (ITH). By comparing the results of baselines and ours, it has been found that our method produces more explainable and understandable results for clinical analysis, which demonstrates the proposed clustering paradigm has the potential with application to knowledge discovery in high dimensional bioinformatics data.

Rational drug combination design in patient-derived avatars reveals effective inhibition of hepatocellular carcinoma with proteasome and CDK inhibitors.

BACKGROUND: Hepatocellular carcinoma (HCC) remains difficult to treat due to limited effective treatment options. While the proteasome inhibitor bortezomib has shown promising preclinical activity in HCC, clinical trials of bortezomib showed no advantage over the standard-of-care treatment sorafenib, highlighting the need for more clinically relevant therapeutic strategies. Here, we propose that rational drug combination design and validation in patient-derived HCC avatar models such as patient-derived xenografts (PDXs) and organoids can improve proteasome inhibitor-based therapeutic efficacy and clinical potential. METHODS: HCC PDXs and the corresponding PDX-derived organoids (PDXOs) were generated from primary patient samples for drug screening and efficacy studies. To identify effective proteasome inhibitor-based drug combinations, we applied a hybrid experimental-computational approach, Quadratic Phenotypic Optimization Platform (QPOP) on a pool of nine drugs comprising proteasome inhibitors, kinase inhibitors and chemotherapy agents. QPOP utilizes small experimental drug response datasets to accurately identify globally optimal drug combinations. RESULTS: Preliminary drug screening highlighted the increased susceptibility of HCC PDXOs towards proteasome inhibitors. Through QPOP, the combination of second-generation proteasome inhibitor ixazomib (Ixa) and CDK inhibitor dinaciclib (Dina) was identified to be effective against HCC. In vitro and in vivo studies demonstrated the synergistic pro-apoptotic and anti-proliferative activity of Ixa + Dina against HCC PDXs and PDXOs. Furthermore, Ixa + Dina outperformed sorafenib in mitigating tumor formation in mice. Mechanistically, increased activation of JNK signaling mediates the combined anti-tumor effects of Ixa + Dina in HCC tumor cells. CONCLUSIONS: Rational drug combination design in patient-derived avatars highlights the therapeutic potential of proteasome and CDK inhibitors and represents a feasible approach towards developing more clinically relevant treatment strategies for HCC.

Clinical consensus statement: Selective internal radiation therapy with yttrium 90 resin microspheres for hepatocellular carcinoma in Asia.

BACKGROUND: Hepatocellular carcinoma (HCC) is subject to different management approaches and guidelines according to Eastern and Western therapeutic algorithms. Use of selective internal radiation therapy (SIRT) with resin yttrium 90 microspheres for HCC has increased in Asia in recent years, without clearly defined indications for its optimal application. The objective of this systematic review and expert consensus statement is to provide guidance and perspectives on the use of SIRT among patients with HCC in Asia. MATERIALS AND METHODS: A systematic literature review identified current publications on HCC management and SIRT recommendations. A group of 10 experts, representing stakeholder specialties and countries, convened between August 2020 and March 2021 and implemented a modified Delphi consensus approach to develop guidelines and indications for use of SIRT for HCC in Asia. Final recommendations were organized and adjudicated based on the level of evidence and strength of recommendation, per approaches outlined by the American College of Cardiology/American Heart Association and Oxford Centre for Evidence-Based Medicine. RESULTS: The experts acknowledged a general lack of evidence relating to use of SIRT in Asia and identified as an unmet need the lack of phase 3 randomized trials comparing clinical outcomes and survival following SIRT versus other therapies for HCC. Through an iterative process, the expert group explored areas of clinical relevance and generated 31 guidance statements and a patient management algorithm that achieved consensus. CONCLUSION: These recommendations aim to support clinicians in their decision-making and to help them identify and treat patients with HCC using SIRT in Asia. The recommendations also highlight areas in which further clinical trials are needed to define the role of SIRT in management of HCC among Asian populations.

Multi-Slice Dense-Sparse Learning for Efficient Liver and Tumor Segmentation.

Accurate automatic liver and tumor segmentation plays a vital role in treatment planning and disease monitoring. Recently, deep convolutional neural network (DCNNs) has obtained tremendous success in 2D and 3D medical image segmentation. However, 2D DCNNs cannot fully leverage the inter-slice information, while 3D DCNNs are computationally expensive and memory intensive. To address these issues, we first propose a novel dense-sparse training flow from a data perspective, in which, densely adjacent slices and sparsely adjacent slices are extracted as inputs for regularizing DCNNs, thereby improving the model performance. Moreover, we design a 2.5D light-weight nnU-Net from a network perspective, in which, depthwise separable convolutions are adopted to improve the efficiency. Extensive experiments on the LiTS dataset have demonstrated the superiority of the proposed method.Clinical relevance- The proposed method can effectively segment livers and tumors from CT scans with low complexity, which can be easily implemented into clinical practice.

Multi-Instance Multi-Label Learning for Gene Mutation Prediction in Hepatocellular Carcinoma.

Gene mutation prediction in hepatocellular carcinoma (HCC) is of great diagnostic and prognostic value for personalized treatments and precision medicine. In this paper, we tackle this problem with multi-instance multi-label learning to address the difficulties on label correlations, label representations, etc. Furthermore, an effective oversampling strategy is applied for data imbalance. Experimental results have shown the superiority of the proposed approach.

The Use of Bilayered Fascia Lata With an Interpositional Omental Flap for Autologous Repair of Contaminated Abdominal Fascial Defects.

INTRODUCTION: Contaminated abdominal fascial defects, such as those seen in enterocutaneous fistula, or wound dehiscence with mesh exposure, are a significant source of morbidity and present unique reconstructive challenges. We present our technique of using the fascia lata, augmented with an interpositional omental flap, for complete autologous reconstruction of contaminated fascial defects, and the postoperative results of 3 cases. METHODS: Three patients with contaminated abdominal defects underwent wound debridement/fistula resection and immediate reconstruction with fascia lata and omentum flap. Defect size ranged from 15 × 8 cm (120 cm) to 25 × 12 cm (300 cm). The fascia lata graft was inset using an underlay technique, and the omentum was tunneled through a subcostal slit in the semilunar line to augment the vascularity of the subcutaneous plane and protect the graft. Skin coverage was achieved by undermining and direct closure or local myocutaneous flaps. RESULTS: Three patients underwent abdominal wall reconstruction with our technique. The median follow-up was 12 months. There were no recurrent infections, fistulae, or herniae. All patients experienced full functional recovery with return to independent activities of daily living by 6 months postoperatively. CONCLUSIONS: Since the use of synthetic material is contraindicated in contaminated abdominal fascial defects. We propose that our combination of fascia lata and an interpositional omental flap is a useful technique for the reconstruction of these challenging defects.

VESPRO: An Individual Patient Data Prospective Meta-Analysis of Selective Internal Radiation Therapy Versus Sorafenib for Advanced, Locally Advanced, or Recurrent Hepatocellular Carcinoma of the SARAH and SIRveNIB Trials.

BACKGROUND: Untreated advanced hepatocellular carcinoma (HCC) has an overall poor prognosis. Currently there are 2 ongoing prospective randomized controlled trials that are evaluating the efficacy and safety of sorafenib and selective internal radiation therapy (SIRT) with yttrium-90 resin microspheres in patients with advanced HCC. The SorAfenib versus Radioembolisation in Advanced Hepatocellular carcinoma (SARAH; 459 patients) trial is being performed in Europe and the SIRt VErsus SorafeNIB (SIRveNIB; 360 patients) trial in the Asia Pacific region. Prospectively combining the results, these trials will not only allow for increased precision to estimate efficacy (in terms of survival), but will also provide increased statistical power for subgroup analyses. OBJECTIVE: To ensure the prospectivity and transparency of the meta-analysis. METHODS: The sirVEnib and SARAH merge PROject (VESPRO) is an individual, patient-data prospective meta-analysis of the SIRveNIB and SARAH randomized trials. The VESPRO protocol includes prespecified hypotheses, inclusion criteria, and outcome measures. The primary outcome measure is overall survival and secondary outcomes include tumor response rate, progression-free survival, progression in the liver as first event, and disease control in the liver. Pooling of toxicity results will allow for robust safety profiles to be established for both therapies, and provides increased statistical power to investigate treatment effects in key subgroups. Analyses will be performed in the intent-to-treat population stratified by trial. RESULTS: Both studies are expected to demonstrate a survival benefit for SIRT together with a better toxicity profile compared with sorafenib. It is also anticipated that liver progression as the first event would be longer in the intervention compared with the control. CONCLUSIONS: As the results of the 2 trials are not yet known, the methodological strength is enhanced, as biases inherent in conventional meta-analyses are avoided. This has the effect of providing this meta-analysis with the advantages of a single, large,randomized study of 819 patients. It is anticipated that the SARAH and SIRveNIB trial results will be published separately and together with the combined meta-analysis results from VESPRO. The combined dataset will allow the effect of the interventions to be explored with improved reliability/precision with respect to prespecified patient and intervention-level characteristics. TRIAL REGISTRATION: Australian New Zealand Trials Registry: ACTRN12617000030370.

Evolution of laparoscopic liver resection at Singapore General Hospital: a nine-year experience of 195 consecutive resections.

INTRODUCTION: We aimed to analyse the changing trends, safety and outcomes associated with the adoption of laparoscopic liver resection (LLR) at a single centre. METHODS: A retrospective review of patients who underwent LLR from 2006 to 2014 at our institution was performed. To explore the evolution of LLR, the study was divided into three equal consecutive time periods (Period 1: 2006-2008, Period 2: 2009-2011, and Period 3: 2012-2014). RESULTS: Among 195 patients who underwent LLR, 24 (12.3%) required open conversions, 68 (34.9%) had resection of tumours in the difficult posterosuperior segments and 12 (6.2%) underwent major (≥ 3 segments) hepatectomies. Median operation time was 210 (range 40-620) minutes and median postoperative stay was 4 (range 1-26) days. Major postoperative morbidity (> Grade II) occurred in 11 (5.6%) patients and 90-day/in-hospital mortality was 1 (0.5%). During the study, the number of LLRs performed showed an increasing trend (Period 1: n = 22; Period 2: n = 19; Period 3: n = 154). Other statistically significant trends were: (a) increase in malignant neoplasms resected; (b) increase in resections of difficult posterosuperior segments; (c) longer median operation time; and (d) decrease in open conversion rates. CONCLUSION: Over the study period, the number of LLRs increased rapidly. LLR was increasingly performed for malignant neoplasms and lesions located in the difficult posterosuperior segments, resulting in longer operation times. However, open conversion rates decreased, and there was no change in postoperative morbidity and other perioperative outcomes.

Personalized predictive lung dosimetry by technetium-99m macroaggregated albumin SPECT/CT for yttrium-90 radioembolization.

BACKGROUND: For yttrium-90 ((90)Y) radioembolization, the common practice of assuming a standard 1,000-g lung mass for predictive dosimetry is fundamentally incongruent with the modern philosophy of personalized medicine. We recently developed a technique of personalized predictive lung dosimetry using technetium-99m ((99m)Tc) macroaggregated albumin (MAA) single photon emission computed tomography with integrated CT (SPECT/CT) of the lung as part of our routine dosimetric protocol for (90)Y radioembolization. Its rationales are the technical superiority of SPECT/CT over planar scintigraphy, ease and convenience of lung auto-segmentation CT densitovolumetry, and dosimetric advantage of patient-specific lung parenchyma masses. METHODS: This is a retrospective study of our pulmonary clinical outcomes and comparison of lung dosimetric accuracy and precision by (99m)Tc MAA SPECT/CT versus conventional planar methodology. (90)Y resin microspheres (SIR-Spheres) were used for radioembolization. Diagnostic CT densitovolumetry was used as a reference for lung parenchyma mass. Pulmonary outcomes were based on follow-up diagnostic CT chest or X-ray. RESULTS: Thirty patients were analyzed. The mean lung parenchyma mass of our Southeast Asian cohort was 822 ± 103 g standard deviation (95% confidence interval 785 to 859 g). Patient-specific lung parenchyma mass estimation by CT densitovolumetry on (99m)Tc MAA SPECT/CT is accurate (bias -21.7 g) and moderately precise (95% limits of agreement -194.6 to +151.2 g). Lung mean radiation absorbed doses calculated by (99m)Tc MAA SPECT/CT and planar methodology are both accurate (bias <0.5 Gy), but (99m)Tc MAA SPECT/CT offers better precision over planar methodology (95% limits of agreement -1.76 to +2.40 Gy versus -3.48 to +3.31 Gy, respectively). None developed radiomicrosphere pneumonitis when treated up to a lung mean radiation absorbed dose of 18 Gy at a median follow-up of 4.4 months. CONCLUSIONS: Personalized predictive lung dosimetry by (99m)Tc MAA SPECT/CT is clinically feasible, safe, and more precise than conventional planar methodology for (90)Y radioembolization radiation planning.

Post-radioembolization yttrium-90 PET/CT - part 1: diagnostic reporting.

BACKGROUND: Yttrium-90 (90Y) positron emission tomography with integrated computed tomography (PET/CT) represents a technological leap from 90Y bremsstrahlung single-photon emission computed tomography with integrated computed tomography (SPECT/CT) by coincidence imaging of low abundance internal pair production. Encouraged by favorable early experiences, we implemented post-radioembolization 90Y PET/CT as an adjunct to 90Y bremsstrahlung SPECT/CT in diagnostic reporting. METHODS: This is a retrospective review of all paired 90Y PET/CT and 90Y bremsstrahlung SPECT/CT scans over a 1-year period. We compared image resolution, ability to confirm technical success, detection of non-target activity, and providing conclusive information about 90Y activity within targeted tumor vascular thrombosis. 90Y resin microspheres were used. 90Y PET/CT was performed on a conventional time-of-flight lutetium-yttrium-oxyorthosilicate scanner with minor modifications to acquisition and reconstruction parameters. Specific findings on 90Y PET/CT were corroborated by 90Y bremsstrahlung SPECT/CT, 99mTc macroaggregated albumin SPECT/CT, follow-up diagnostic imaging or review of clinical records. RESULTS: Diagnostic reporting recommendations were developed from our collective experience across 44 paired scans. Emphasis on the continuity of care improved overall diagnostic accuracy and reporting confidence of the operator. With proper technique, the presence of background noise did not pose a problem for diagnostic reporting. A counter-intuitive but effective technique of detecting non-target activity is proposed, based on the pattern of activity and its relation to underlying anatomy, instead of its visual intensity. In a sub-analysis of 23 patients with a median follow-up of 5.4 months, 90Y PET/CT consistently outperformed 90Y bremsstrahlung SPECT/CT in all aspects of qualitative analysis, including assessment for non-target activity and tumor vascular thrombosis. Parts of viscera closely adjacent to the liver remain challenging for non-target activity detection, compounded by a tendency for mis-registration. CONCLUSIONS: Adherence to proper diagnostic reporting technique and emphasis on continuity of care are vital to the clinical utility of post-radioembolization 90Y PET/CT. 90Y PET/CT is superior to 90Y bremsstrahlung SPECT/CT for the assessment of target and non-target activity.

Post-radioembolization yttrium-90 PET/CT - part 2: dose-response and tumor predictive dosimetry for resin microspheres.

BACKGROUND: Coincidence imaging of low-abundance yttrium-90 (90Y) internal pair production by positron emission tomography with integrated computed tomography (PET/CT) achieves high-resolution imaging of post-radioembolization microsphere biodistribution. Part 2 analyzes tumor and non-target tissue dose-response by 90Y PET quantification and evaluates the accuracy of tumor 99mTc macroaggregated albumin (MAA) single-photon emission computed tomography with integrated CT (SPECT/CT) predictive dosimetry. METHODS: Retrospective dose quantification of 90Y resin microspheres was performed on the same 23-patient data set in part 1. Phantom studies were performed to assure quantitative accuracy of our time-of-flight lutetium-yttrium-oxyorthosilicate system. Dose-responses were analyzed using 90Y dose-volume histograms (DVHs) by PET voxel dosimetry or mean absorbed doses by Medical Internal Radiation Dose macrodosimetry, correlated to follow-up imaging or clinical findings. Intended tumor mean doses by predictive dosimetry were compared to doses by 90Y PET. RESULTS: Phantom studies demonstrated near-perfect detector linearity and high tumor quantitative accuracy. For hepatocellular carcinomas, complete responses were generally achieved at D70 > 100 Gy (D70, minimum dose to 70% tumor volume), whereas incomplete responses were generally at D70 < 100 Gy; smaller tumors (<80 cm3) achieved D70 > 100 Gy more easily than larger tumors. There was complete response in a cholangiocarcinoma at D70 90 Gy and partial response in an adrenal gastrointestinal stromal tumor metastasis at D70 53 Gy. In two patients, a mean dose of 18 Gy to the stomach was asymptomatic, 49 Gy caused gastritis, 65 Gy caused ulceration, and 53 Gy caused duodenitis. In one patient, a bilateral kidney mean dose of 9 Gy (V20 8%) did not cause clinically relevant nephrotoxicity. Under near-ideal dosimetric conditions, there was excellent correlation between intended tumor mean doses by predictive dosimetry and those by 90Y PET, with a low median relative error of +3.8% (95% confidence interval, -1.2% to +13.2%). CONCLUSIONS: Tumor and non-target tissue absorbed dose quantification by 90Y PET is accurate and yields radiobiologically meaningful dose-response information to guide adjuvant or mitigative action. Tumor 99mTc MAA SPECT/CT predictive dosimetry is feasible. 90Y DVHs may guide future techniques in predictive dosimetry.

Lipoma of the pancreas, a case report and a review of the literature.

Lipomas of the pancreas are very rare. There are fewer than 25 reported cases of lipoma originating from the pancreas. We present a case of pancreatic lipoma in a 61-year-old woman with magnetic resonance imaging findings and confirmatory histological findings. We discuss and highlight the radiological features distinguishing a pancreatic lipoma from other fatty lesions of the pancreas and pancreatic liposarcoma and provide a brief review of the literature.

The use of submicron/nanoscale PLGA implants to deliver paclitaxel with enhanced pharmacokinetics and therapeutic efficacy in intracranial glioblastoma in mice.

Pharmacokinetics and therapeutic efficacy of submicron/nanoscale, intracranial implants were evaluated for treating malignant glioblastoma in mice. 9.1% (w/w) paclitaxel-loaded polylactide-co-glycolide (PLGA) nanofiber discs (F3) were fabricated and characterized for morphology and size distribution. Along with F3, three other formulations, 9.1% (w/w) paclitaxel-loaded PLGA submicron-fiber discs (F2), 16.7% (w/w) paclitaxel-loaded PLGA microspheres entrapped in hydrogel matrices (H80 and M80) were intracranially implanted in BALB/c mice and the coronal brain sections were analyzed for bio-distribution of paclitaxel on 14, 28 and 42 days post-implantation. BALB/c nude mice with intracranial human glioblastoma (U87 MG-luc2) were used in the therapeutic efficacy study. Animals were randomized to intracranial implantation of F3 and H80 with paclitaxel dose of 10mg/kg, placebo F3, placebo H80, weekly intratumoral injection of Taxol (10mg/kg) or no treatment and the treatment response was analyzed by bioluminescence imaging and histological (H&E, Ki-67) examinations. Enhanced, therapeutic paclitaxel penetration (approximately 1 microm) in the mouse brain up to 5mm from the implant site even after 42 days post-implantation from F3 and H80 was confirmed and deduced to be diffusion/elimination controlled. F3 and H80 demonstrated significant (approximately 30 fold) tumor inhibition and significantly low tumor proliferation index after 41 days of treatment in comparison to sham and placebo controls. The submicron/nanoscale implants are able to demonstrate optimal paclitaxel pharmacokinetics in the brain/tumor with significant tumor inhibition in a glioblastoma xenograft model in mice and hence could be potentially useful to treat highly recurrent GBM.

Identification of proteins differentially expressed between capillary endothelial cells of hepatocellular carcinoma and normal liver in an orthotopic rat tumor model using 2-D DIGE.

Hepatocellular carcinoma (HCC) is one of the deadliest cancers with few treatment options. It is a hypervascular tumor in which angiogenesis plays a critical role in its progression. Tumor capillary endothelial cells (TECs) in HCC are known to originate from liver sinusoid endothelial cells, which then go through a capillarization process to become morphologically as well as functionally different TECs. In this work, we investigated proteins differentially expressed between freshly isolated TECs and sinusoid endothelial cells from well-formed rat HCC using 2-D DIGE coupled with MALDI-TOF/TOF MS. Thirty-eight unique proteins were identified to be differentially expressed more than twofold between the two endothelial cell types. Amongst the differentially expressed proteins, two novel endothelial markers, EH domain-containing protein 3 and galectin-3, were confirmed by Western blot and immunohistochemistry in both rat and human HCC samples. We showed that EH domain-containing protein 3 is significantly down-regulated in TECs, but galectin-3 is up-regulated. We propose possible roles of these two proteins in tumor vessel development in HCC.

Neoadjuvant and adjuvant therapy for surgical resection of hepatocellular carcinoma.

BACKGROUND: Hepatocellular carcinoma is a disease of great concern. Surgery is the treatment of choice, but there is still a high recurrence rate after resection. OBJECTIVES: To determine the benefits and harms of neoadjuvant and adjuvant therapies compared to surgery alone or surgery and placebo/supportive therapy after curative resection for operable hepatocellular carcinoma. SEARCH STRATEGY: We searched The Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded, Chinese Biomedical Database, and US National Cancer Institute's Physician's Data Query Trials Database until 2005. References of the identified trials were also searched for identifying further trials. SELECTION CRITERIA: Randomised and quasi-randomised trials that compared hepatocellular carcinoma patients who were given and not given neoadjuvant/adjuvant therapy as a supplement to curative liver resection. DATA COLLECTION AND ANALYSIS: Data were extracted independently by two authors and discrepancies resolved by consensus. The survival and disease-free survival curves were compared using their one, two, three, four, and five-year survival rates, median survival times, and the result of the significance tests (P-values). MAIN RESULTS: A total of 12 randomised trials were identified, totaling 843 patients. The size of the randomised clinical trials ranged from 30 to 155 patients. Both preoperative (neoadjuvant) and postoperative (adjuvant), systemic and locoregional (+/- embolisation), chemo- and immunotherapy interventions were tested. Treatment regimens and patients selected were not comparable, so no pooling was done. Only one regimen using preoperative transcatheter arterial chemoembolisation with doxorubicin was similar in two trials. Four of the twelve trials reported survival benefit at five years when given adjuvant or neoadjuvant therapy. Disease-free survival was reported in nine trials, and the estimated hazard ratios show that disease-free survival was significant in two trials at five years. These two trials had not shown a survival advantage, but the recurrence was significantly lower in patients given adjuvant or neoadjuvant therapy. The highest toxicity rate was in a trial using oral 1-hexylcarbamoyl 5-fluorouracil which resulted in 12 out of 38 patients being withdrawn from the trial because of adverse events. AUTHORS' CONCLUSIONS: There is no clear evidence for efficacy of any of the adjuvant and neo-adjuvant protocols reviewed, but there is some evidence to suggest that adjuvant therapy may be beneficial offering prolonged disease-free survival. In order to detect a realistic treatment advantage, larger trials with lower risk of systematic error will have to be conducted.

Adult choledochal cysts: an audit of surgical management.

BACKGROUND: Choledochal cysts are rare congenital cystic dilatations of the biliary tree. Surgical management has evolved with regards to timing and surgical procedure of choice. We conducted a retrospective review of clinical presentation and surgical management of adult choledochal cysts. METHODS: Thirty-two patients with choledochal cysts who underwent surgery between April 1991 and January 2005 were reviewed. There were 27 Todani Type I, 2 Type II, 2 Type IVA and 1 Type V cysts. Eighty-four per cent of patients underwent complete cystectomy and hepaticojejunostomy. Seven patients had revision surgery comprising completion cystectomy and hepaticojejunostomy. RESULTS: There were no perioperative mortalities. Perioperative morbidity rate was 44% and the commonest complication perioperatively was wound infection (19%). Malignancy was noted in one histological specimen. This patient was disease free for 1 year postoperatively and was subsequently lost to follow up. No further malignancy was found on median follow up of 3.9 years (range, 1-14 years) for the other 31 patients. CONCLUSION: Adult choledochal cysts are rare and are often non-specific in their clinical presentation. In managing patients with choledochal cysts, it is important to first treat complications such as sepsis and pancreatitis before imaging of the biliary tree with endoscopic retrograde cholangiopancreatography or magnetic resonance cholangiopancreatography to evaluate the full extent and type of choledochal cyst. Surgical management should be planned single-stage surgery comprising complete cyst resection, cholecystectomy and Roux-en-Y hepaticojejunostomy and should be carried out by hepatobiliary specialists. Excellent perioperative morbidity and mortality results are possible with this strategy. Malignancy is rare and was only noted in 3% but close follow up is warranted.

Factors affecting early mortality in spontaneous rupture of hepatocellular carcinoma.

BACKGROUND: Spontaneous rupture of hepatocellular carcinoma (HCC) is a catastrophic surgical emergency with high mortality rates. The aim of this study is to determine the factors associated with the prognosis and to assess the outcome of different management strategies. METHODS: A retrospective study of 34 consecutive patients with spontaneous rupture of HCC was conducted from January 1996 to January 2004. Clinical, biochemical and operative factors influencing 30-day mortality were analysed. RESULTS: In our study, 30-day mortality rate was 32% (n = 11). Presence of cirrhosis, Child's C status, shock on admission, higher blood transfusion requirement, raised alpha-fetoprotein, raised alkaline phosphatase, raised aspartate transaminase, and raised indocyanine green at 15 min were all associated with increased risk of 30-day mortality on univariate analysis (P < 0.05). On multivariate analysis, only shock on admission (P = 0.001) and higher blood transfusion requirement (P = 0.01) were significant independent factors affecting early mortality. Surgical intervention was associated with a better 30-day survival as compared with medical therapy or transarterial embolization. The median survival time of patients undergoing curative resection was significantly longer than that of patients who had surgery for haemostasis only (420 vs 205 days). The overall median survival was 161 days. CONCLUSIONS: Spontaneous rupture of HCC is a potentially salvageable complication of HCC. Poor prognosis is associated with poor liver reserve, advanced disease and severity of haemorrhage. Shock and blood transfusion requirement are the only independent factors affecting early mortality.