Examination of Timing of Palliative Care Consultation in Patients With Grade 3 Acute on Chronic Liver Failure at a Large Academic Transplant Center

Introduction: Patients with acute on chronic liver failure (ACLF) experience high morbidity and mortality with a decreased quality of life. High quality palliative care (PC) along with extensive goals of care (GOC) conversations may be beneficial in this population. Methods: Retrospective chart review of individuals with cirrhosis admitted to the ICU at 2 large referral centers. Appropriate International Classification of Diseases-10th Edition (ICD-10) codes for cirrhosis were used to screen patients with an intensive care unit (ICU) ICU admission from 1/1/2016-12/31/2019 and confirmed through manual chart review. The primary outcome was impact of timing of PC consultation in severe acute on chronic liver failure on GOC conversations and health care utilization. Results: 1,103 patients were included in the study, 285 (26%) were identified as having ACLF grade 3, and only 62 (22%) received a PC consultation. Patients were grouped based on timing of PC consultation. Early consultation was defined as initial consult ≤72 hours (n=22), and late consultation was defined as >72 hours (n=40) from admission. There was no difference in the age between the 2 groups; however, there were more males (77%) in the early consult group compared to 48% in the late consult group (P=0.023). There was no difference in etiology of cirrhosis between the 2 groups, with alcohol related cirrhosis being the most common. In the early consult group, GOC conversations occurred sooner with a median time to first GOC conversation of 1 day compared to 7 days in the late consult group (P< 0.001). Despite an earlier consult, the total number of GOC conversations were similar between the 2 groups with a median of 3 conversations. For health care utilization, there was no difference in initiation of dialysis (77% vs 63%) or mechanical ventilation (77% vs 68%), vasopressor use (86% vs 88%) or ICU readmission (0% vs 10%), between the early and late palliative care consultation groups, respectively. Overall outcomes were similar in terms of ICU death (55% vs 58%), in-hospital non-ICU death (9% vs 13%), discharge to hospice (14% vs 15%), discharged to home/SNF (23% vs 15%) and 30-day mortality (82% vs 78%) between the early and late consultation groups, respectively. Conclusion: Despite high in hospital and 30-day mortality, PC is underutilized in patients with grade 3 ACLF; however, when PC is involved, timing does not change outcomes. Efforts to increase high quality palliative care in this population is warranted (Table 1). Table 1. - Comparison of Baseline Characteristics and Palliative Care Utilization Between Early and Late Palliative Consultation Groups Characteristic PC consult ≤3 days (N=22) PC consult ≥4 days (N=40) P-value Age (SD) 52.0 (16.8) 54.4 (9.9) 0.475 Male Gender, n (%) 17 (77.3) 19 (47.5) 0.023 Race, n (%) Black of African American 4 (18.2) 3 (7.5) 0.287 White 15(68.2) 34 (85.0) Other 3 (13.6) 3 (7.5) Cirrhosis etiology, n (%) HCV 1 (4.5) 7 (17.5) 0.491 HCV +EtOH 3 (13.6) 3 (7.5) ALD 9 (40.9) 15 (37.5) NASH 3 (13.6) 8 (20) Other 6 (27.3) 7 (17.5) Health care utilization, n (%) Initiation of dialysis 17 (77.3) 25(62.5) 0.234 Required mechanical ventilation 17 (77.3) 27 (67.5) 0.417 Vasopressor use 19 (86.4) 35 (87.5) 0.898 ICU readmission 0 (0.0) 4 (10.0) 0.287 # of GOC conversations, median (IQR) 3 (2, 4) 3 (2, 4) 0.647 Time to GOC, median (IQR) 1 (0, 2) 7 (3, 14) < 0.001 Trigger for GOC, n (%) Upgrade to ICU 9 (40.9) 8 (20.0) 0.077 Bleed 1 (4.5) 4 (10.0) 0.647 Change in respiratory status 10 (45.5) 19 (47.5) 0.877 Need for dialysis 7 (31.8) 10 (25.0) 0.568 Hypotension/vasopressor need 12 (54.5) 20 (50.0) 0.732 Primary team requested 6 (27.3) 15 (37.5) 0.576 Family requested 2 (9.1) 2 (5.0) 0.610 Other 7 (31.8) 6 (15.0) 0.120 Involved in GOC, n (%) Patient only 0 (0.0) 3 (7.5) 0.406 Patient and family 9 (40.9) 14 (35.0) Family/friend only 13 (59.1) 23 (57.5) Why patient could not participate, n (%) Intubated 8 (36.4) 16 (40.0) 1.000 Metabolic encephalopathy 8 (36.4) 16 (40.0) 1.000 Other 7 (31.8) 6 (15.0) 0.120 Final disposition, n (%) ICU death 12 (54.5) 23 (57.5) 0.822 In-hospital, non-ICU death 2 (9.1) 5 (12.5) 1.000 Discharged to hospice 3 (13.6) 6 (15.0) 1.000 Discharged to home/SNF 5 (22.7) 6 (15.0) 0.499 30-day mortality (%) 81.8 77.5 0.566