Faculty Bibliography

BACKGROUND AND OBJECTIVES/OBJECTIVE:While preoperative treatment is frequently administered to CRLM patients, the impact of chemotherapy, with or without bevacizumab, on liver regeneration remains controversial. METHODS:The early and late regeneration indexes were defined as the relative increase in liver volume (RLV) within 2 and 9 months from surgery. Regeneration rates of the preoperative treatment groups were compared. RESULTS:Preoperative chemotherapy details and volumetric data were available for 185 patients; 78 (42.2%) received preoperative chemotherapy with bevacizumab (Bev+), 46 (24.8%) received chemotherapy only (Bev-), and 61 (33%) received no chemotherapy. Patients in the Bev+ and Bev- groups received similar chemotherapy cycles (4 [3-6] vs 4 [4-6]; P = 0.499). Despite the comparable clinicopathological characteristics and Resected Volume/Total Liver Volume (TLV) at surgery (P = 0.944) of both groups, Bev+ group had higher early and late regeneration (17.2% vs 4.3%; P = 0.035 and 14.0% vs 9.4%; P = 0.091, respectively). Of note, early and late regeneration rates (3.7% and 10.9% vs 6.6% and 5.5%, respectively) were comparable between the no chemotherapy and Bev- groups (all P > 0.05). In multivariable analysis -adjusted for gender, age, portal vein embolization, preoperative chemotherapy, resected liver volume, tumor number, postoperative chemotherapy, fibrosis, steatosis- bevacizumab independently predicted early liver regeneration (P = 0.019). CONCLUSION/CONCLUSIONS:Our findings suggest that preoperative bevacizumab administered along with chemotherapy was associated with enhanced volumetric restoration. Interestingly, this effect was more pronounced among patients who received oxaliplatin-based regimens and bevacizumab compared to those treated with irinotecan-based regimens and bevacizumab.

Background and Objectives Despite routine inclusion of lymphovascular invasion (LVI) status in pathologic reports of resected pancreatic ductal adenocarcinomas (PDA), the clinical implications of LVI have not been well characterized. Methods This study is a retrospective review of 2640 patients who underwent a pancreatectomy for PDA at Thomas Jefferson University Hospital, Massachusetts General Hospital, or Johns Hopkins Hospital (2003-2014). Clinical and pathologic records were extracted from institutional databases. Results The median post-resection survival for the total cohort was 19.2 months with a 5-year survival rate of 15.2%. In a multivariate Cox proportional hazards model including conventional pathologic features, LVI was an independent predictor of survival (HR = 1.14, P = 0.017). In a stratified Kaplan-Meier survival analysis, patients with N0, LVI- PDA had a significantly improved overall survival compared to those with N0, LVI+ PDA (median 31 vs 24 mo, P = 0.020). Similarly, patients with N1, LVI- PDA had superior survival to patients with N1, LVI+ disease (18.6 vs 16.5 mo, P = 0.001). Conclusions As the first large scale study focused on the clinical impact of LVI status in PDA, these data indicate that this routinely reported pathologic feature is a bona fide and independent adverse prognostic factor.

BACKGROUND:Although laparoscopic distal pancreatectomy is considered a standard approach, 10% to 40% of these are converted. The preoperative risk factors for conversion are not well described. The aim of this study was to identify risk factors associated with conversion. METHODS:Clinicopathological variables of 211 consecutive patients who underwent laparoscopic distal pancreatectomy between January 2007 and December 2015 at Johns Hopkins were analyzed to identify factors associated with conversion. Furthermore, the learning curve for laparoscopic distal pancreatectomy was studied. RESULTS:On univariate analysis of diabetes mellitus, preoperative diagnosis of malignant disease, multiorgan resection, surgeons' years and case experience were significantly associated with conversion (all P < .05). Risk factors independently associated with conversion included diagnosis of malignant disease (odds ratio = 5.40; 95% confidence interval, 1.93-15.12, P = .001), multiorgan resection (odds ratio = 7.10; 95% confidence interval, 1.60-31.53, P = .01), and surgeons' case experience (odds ratio = 0.32; 95% confidence interval, 0.12-0.85, P = .023). Intraoperative reasons for conversion included presence of excessive intraabdominal and retroperitoneal fat (N = 10, 32.3%), adhesions (N = 10, 32.3%), extent of tumor invasion (N = 8, 25.8%), anatomy of vessels (N = 6, 19.4%), and intraoperative bleeding (N = 2, 6.5%). CONCLUSION:Patients undergoing laparoscopic distal pancreatectomy with a preoperative diagnosis of malignant disease or possible multiorgan resection are at a higher risk of conversion. Surgeon experience of performing >15 procedures significantly reduces the risk of conversion.