Prophylactic transfusions in pediatric patients with liver disease: Effect of guidelines on tertiary center practice

Abstract

Introduction: Advanced liver disease encompasses acute liver failure, chronic compensated cirrhosis and end-stage liver disease. These patients often have an increased INR (international normalized ratio), reflecting poor liver synthetic function, which can lead to a misperception of them being "coagulopathic." However, it has been demonstrated that INR is not predictive of excessive bleeding in minor procedures in those with liver disease. In everyday practice, these patients are often given "prophylactic" blood product transfusions before procedures. Fresh frozen plasma (FFP) is often administered to coagulopathic patients who are not bleeding, despite it not reliably correcting coagulopathy in liver disease and running the risk of increasing portal pressure. Furthermore, there is increasing awareness that the reduction of both procoagulant and anticoagulant factors leads to a rebalanced hemostasis. With this evidence in mind, our tertiary pediatric center introduced guidelines for blood product transfusions in invasive procedures in children with advanced liver disease. We evaluate the effect of these guidelines in relation to clinical practice and costs over two time periods. Methods: We undertook a retrospective review of all patients with advanced liver disease who underwent procedures and received prophylactic transfusions from November 2011 to January 2017. The cohort was di-vided into two periods: pre-guidelines (C1) and post-guidelines (C2). Results: Data were collected for 29 children who underwent 82 procedures: 54 in C1 and 28 in C2. Liver disease was predominantly metabolic liver disease (33%), biliary atresia (28%) and autoimmune hepatitis (21%). When comparing C1 with C2, frequency of transfusion decreased significantly (85% vs 57%, P = 0.02). This decrease in frequency was observed for individual products as well: FFP (57% vs 44%), cryoprecipitate (39% vs 25%) and platelets (65% vs 50%). The most frequent procedures were central line insertion (30% vs 25%), endoscopy for variceal surveillance (14% vs 12%), central line removal (4% vs 6%), open liver biopsy (9% vs 0), and percutaneous biopsy (15% vs 19%). Mean units of products used in C2 decreased significantly in endoscopy for variceal surveillance (3.4 vs 1, P = 0.005), liver biopsies (1.08 vs 0.4, P = 0.001) and central line insertion or removal (2.8 vs 0.6, P = 0.001). Of those with an INR >1.5, the number of procedures performed decreased between C1 and C2 (83% vs 63%), but prophylactic FFP was administered with similar frequency (61% vs 60%). Improvement in INR after FFP administration was limited, with a mean decrease of 0.1 in both periods. Total transfusion costs decreased in C2 compared with C1 ($41 237.58 vs $9189.07); similar findings were noted with total nursing costs ($8134.29 vs $1927.82). Conclusions: Frequency of prophylactic blood transfusions in those with advanced liver disease undergoing procedures has reduced over time with implementation of hospital guidelines. This has promising implications for improved cost savings and decreased hospital length of stay. While transfusions of FFP have decreased overall within our cohort, they are still frequently used in coagulopathic patients, with minimal change in INR. Our guidelines, while promoting a promising trend in reduction of transfusions, are recent, and ongoing education will be required to increase awareness that the patient with liver disease has altered hemostasis and prophylactic transfusions have limited effect in the absence of active bleeding.L6244311782018-10-23 <br />