871 Liver Transplant for Colorectal Liver Metastases: A Systematic Review and Meta-Analysis of Proportions

Introduction

Colorectal liver metastases were historically considered a contraindication to liver transplantation, but dismal outcomes for those with metastatic colorectal cancer and advancements in liver transplantation (LT) have led to a renewed interest in the topic. We aim to compare the current evidence for liver transplantation for non-resectable colorectal liver metastases (NRCLM) with the current standard treatment of palliative chemotherapy reported in literature – 5-year survival rate <10%.

Method

A systematic review and meta-analysis of proportions was conducted following screening of MEDLINE, EMBASE, SCOPUS and CENTRAL for studies reporting liver transplantation for colorectal liver metastases. Post-operative outcomes measured included 1-, 3- and 5-year survival, overall survival, disease-free survival, and complication rates.

Results

Three non-randomised studies met the inclusion criteria, reporting a total of 48 patients receiving LT for NRCLM. Survival at 1-, 3- and 5-years was 83.3-100%, 58.3-80% and 50-80% respectively with no significant difference detected (p = 0.22, p = 0.48, p = 0.26). Disease free survival was 35-56% with the most common site of recurrence being lung. Thirteen out of fourteen deaths were due to disease recurrence.

Conclusions

Although current evidence suggests a survival benefit conferred by LT in NRCLM compared to palliative chemotherapy, the ethical implications of organ availability and allocation demand rigorous justification. Eight registered clinical trials will report on 300 more patients undergoing LT for NRCLM over the next 10 years: concomitant improvements in the management of patients following liver resection and of palliative chemotherapy regimens is paramount.

715 The Importance of Social Media to The Academic Surgical Literature: Relationship Between Twitter Activity and Readership Metrics

Aim

Social media (SoMe) has an increasing role within professional surgical practice, including the publishing and engagement of academic literature. This study aims to analyse the relationship between social media use and traditional and alternative metrics amongst academic surgical journals.

Method

Journals were identified through the InCites Journal Citation Reports 2019, and their impact factor (IF), h-index and CiteScore were noted. Social media platforms were examined, and Twitter activity interrogated between 1st January- 31st December 2019. Healthcare Social Graph (HSG) score and an aggregated Altmetric score were also calculated for each journal. Statistical analysis was carried out to look at the correlation between traditional metrics, Twitter activity and altmetrics.

Results

Journals with higher IF were more likely to use a greater number of SoMe platforms (R2=0.648; p < 0.0001). Journals with dedicated Twitter profiles had a higher IF than journals without (median, 2.96 vs 1.88; MWU=390; p < 0.001) however over a one-year period (2018-2019) having a twitter presence did not alter IF (MWU=744.5; p = 0.885). Increased Twitter activity was positively correlated with IF. Longitudinal analysis over six years suggested cumulative tweets correlated with an increased IF (R2=0.324, p = 0.004). Novel alternative measures including HSG score (R2=0.472, p = 0.005) and Altmetric score (R2=0.779, p = 0.001) positively correlated with IF.

Conclusions

Higher IF is associated with SoMe presence and activity, particularly on Twitter, with long term activity being of particular importance. Modern alternative metrics correlate with IF. This relationship is complex and future studies should look to understand this further.

720 Pringle Versus Selective Hepatic Vascular Exclusion in Major Hepatectomy: A Systematic Review and Meta-Analysis

Aim

Mortality and morbidity following hepatic resection is significantly affected by major intra-operative blood loss. Multiple techniques have been developed in an attempt to minimise blood loss by occluding hepatic inflow and outflow. This systematic review and meta-analysis evaluates whether selective hepatic vascular exclusion compared to a Pringle manoeuvre in hepatic resection reduces rates of morbidity and mortality.

Method

A systematic review and meta-analysis were conducted according to the PRISMA guidelines by screening EMBASE, MEDLINE/PubMed, CENTRAL, SCOPUS and bibliographic reference lists for comparative studies meeting the predetermined inclusion criteria. Intra- and post-operative outcome measures were investigated. Pooled odds ratios or mean differences with 95% confidence intervals were calculated using either fixed- or random-effects models.

Results

Five studies were identified including two randomized controlled trials and three observational studies reporting a total of 2,198 patients. Data synthesis showed significantly decreased rates of mortality, overall complications, patients requiring blood transfusion, air embolism, warm ischaemia time, liver failure and multi-organ failure when performing SHVE compared to a Pringle manoeuvre. Rates of hepatic vein rupture and post-operative haemorrhage remained the same. Performing SHVE resulted in a significantly longer operation time.

Conclusions

Performing SHVE in major hepatectomy may result in reduced rates of morbidity and mortality when compared to a Pringle manoeuvre, although may prolong operating time. The results of this meta-analysis are based on a few high-quality studies where tumours were adjacent to major vessels. Further RCTs are required to validate these results and determine the best technique for hepatic vascular control in this patient cohort.

717 Roux-En-Y Versus Single Loop Reconstruction in Pancreaticoduodenectomy: A Systematic Review and Meta-Analysis

Aim

Post-operative pancreatic fistula (POPF) and delayed gastric emptying (DGE) both remain problematic complications following pancreaticoduodenectomy. This systematic review and meta-analysis evaluates whether Roux-en-Y compared to a single loop reconstruction in pancreaticoduodenectomy significantly reduces rates of these complications.

Method

A systematic review and meta-analysis were conducted according to the PRISMA guidelines by screening EMBASE, MEDLINE/PubMed, CENTRAL and bibliographic reference lists for comparative studies meeting the predetermined inclusion criteria. Post-operative outcome measures included: POPF, DGE, bile leak, operating time, blood loss, need for transfusion, wound infection, intra-abdominal collection, post-pancreatectomy haemorrhage, overall morbidity, re-operation, overall mortality, hospital length of stay. Pooled odds ratios or mean differences with 95% confidence intervals were calculated using either fixed- or random-effects models.

Results

Fourteen studies were identified including four randomized controlled trials (RCTs) and 10 observational studies reporting a total of 2 031 patients. Data synthesis showed no statistically significant difference between the two groups in any of the outcome measures except operating time, which was longer in those undergoing Roux-en-Y reconstruction.

Conclusions

Roux-en-Y is not superior to single loop reconstruction in pancreaticoduodenectomy but may prolong operating time. Future high-quality randomized studies with appropriate study design and sample size power calculation may be required to further validate this conclusion.

Treatment of lithium induced tremor with atenolol

This is the first report on the successful treatment of one patient with lithium induced tremor with hydrophilic atenolol, which is a relatively selective beta 1 adrenergic receptor blocker. Atenolol's advantages over lipophilic beta blockers in the treatment of lithium induced tremor are discussed.