The role of bile duct brush cytology in distinguishing malignant from benign biliary obstruction in patients stratified by clinical parameters for risk of malignancy, a retrospective review

Abstract

NSQIP) database and hepatectomy data-file were analyzed. MIH performed without planned conversion was compared to those requiring unplanned conversion. Results: Among 3064 hepatectomy patients, MIH was performed in 549 (17.9%); 520 (94.7%) laparoscopic and 29 (5.3%) robotic. Resection was performed for metastases in 225 (41%), benign lesions in 178 (32%), primary hepatic malignancy in 130 (24%) and unknown diagnosis in 16 (3%) patients. Major hepatectomy (!3 segments) was performed in 91 (16.6%). Unplanned conversion was required in 115 (20.9%). On multivariate analysis, hypertension (OR 2.27, p = 0.0004), concurrent ablation (OR 2.79, p = 0.0028), cirrhotic liver texture (OR 1.97, p = 0.0306), Pringle maneuver (OR 3.47, p < 0.0001), and major hepatectomy (OR 2.61, p = 0.0009) were significantly associated with unplanned conversion. Patients with unplanned conversion experienced higher rates of bile leak (8.7% vs 2.5%, p = 0.0049), wound dehiscence (1.7% vs 0.0%, p = 0.0436), UTI (6.1% vs 0.9%, p = 0.0023) and perioperative transfusion (25.2% vs 6.7%, p < 0.0001). Unplanned conversion was also associated with greater LOS (7.2 AE 6.9 vs 4.0 AE 5.1 days, p < 0.0001) and 30-day mortality (3.5% vs 0.7%, p = 0.0381). Conclusion: Analysis of this large national database revealed unplanned conversion during MIH is associated with significantly higher morbidity and mortality. Furthermore, this investigation identified several factors that should be carefully considered when selecting patients for MIH.

Diagnostic sensitivity and risk of malignancy for bile duct brushings categorized by the Papanicolaou Society of Cytopathology System for reporting pancreaticobiliary cytopathology

“…Diagnostic sensitivity for bile duct brushings has been variable with reported sensitivities ranging from 40% to 60% in the majority of studies. [10][11][12][13][14][15][16][17] Our finding of a 64% sensitivity is similar. A number of authors have attempted to define diagnostic criteria which would improve overall sensitivity with many of these updated criteria demonstrating a sensitivity of between 35% and 83%.…”