Fifteen-Year Single-Center Experience of Biliary Complications in Liver Trauma Patients: Changes in the Management of Posttraumatic Bile Leak.

INTRODUCTION: Management of posttraumatic bile leak has evolved over time in our unit, from endoscopic retrograde cholangiopancreatography (ERCP) stenting to intraperitoneal drainage (IPD) alone as first-line treatment for intraperitoneal bile leak. MATERIALS AND METHODS: Retrospective review of liver trauma patients from 2002 to 2017. Demographics, time and mode of diagnosis of bile leak, management, and outcome were analyzed of the box plot. RESULTS: In 118 patients, there were 28 traumatic bile leaks. Eighteen were free intraperitoneal and 10 were localized bilomas. The median time of diagnosis was 6 days following injury. The modes of diagnosis were preemptive hepatobiliary scintigraphy (18), computed tomography (CT) or ultrasound (7), and laparotomy (3). Free intraperitoneal biliary leak management included 11 IPD alone, 3 IPD plus ERCP, 2 IPD plus transcystic biliary stent (TBS), 1 operative cholangiogram, and 1 no intervention. Median time of IPD duration was 7 days (4-95) in IPD alone versus 14 days (6-40) in IPD + ERCP/TBS (p = 0.3). Median inpatient length of stay was 13 days (8-44) in IPD alone versus 12 days (8-22) in IPD + ERCP/TBS (p = 0.4). CONCLUSION: Placement of IPD alone, as first-line treatment, is safe and effective in the management of intraperitoneal bile leaks, avoiding the costs and potential complications of ERCP.

Hepatopulmonary Syndrome in Children: A 20-Year Review of Presenting Symptoms, Clinical Progression, and Transplant Outcome.

Hepatopulmonary syndrome (HPS) in stable patients with cirrhosis can easily be overlooked. We report on the presenting symptoms, disease progression, and outcomes after liver transplantation (LT) in children with HPS. Twenty patients were diagnosed with HPS between 1996 and 2016. The etiologies were as follows: biliary atresia (n = 9); alpha-1-antitrypsin deficiency (n = 2); cryptogenic liver disease (n = 3); and others (n = 6). HPS presentations were as follows; dyspnea (n = 17) and pneumonia (n = 3). For diagnostic confirmation, the following techniques were used: technetium-99m-labeled macroaggregated albumin lung perfusion scan (n = 13) or contrast echocardiogram (n = 7). There were 16 patients listed for LT, with a median age at HPS diagnosis of 10 years and an average wait from listing to LT of 9 weeks. A marked rise in hemoglobin (Hb; median, 125-143.5 g/L) and modest decrease in oxygen saturation (SpO2 ; median 91% to 88% room air) were evident over this time. Patients' need for assisted ventilation (1 day), pediatric intensive care unit (PICU) stay (3 days), and total hospital stay (20 days) were similar to our general LT recipients-the key difference in the postoperative period was the duration of supplementary O2 requirement. Hb of ≥130 g/L on the day of LT correlated with a longer PICU stay (P value = 0.02), duration of supplementary O2 (P value = 0.005), and the need for the latter beyond 7 days after LT (P value = 0.01). Fifteen patients had resolution of their HPS after LT. The 5-, 10-, and 20-year survival rates were unchanged at 87.5%. None had a recurrence of HPS. In conclusion, HPS is a life-threatening complication of cirrhosis which usually develops insidiously. This combined with the often-stable nature of the liver disease leads to delays in diagnosis and listing for LT. Progressive polycythemia extends the need for supplementary O2 and PICU stay. We advocate screening for HPS with a combination of SpO2 and Hb monitoring to facilitate earlier recognition, timely LT, and shortened recovery periods.