886. Non-recognized Liver Impairment in Infected Critically Ill Patients is Frequent and Hazardous
Session: Poster Abstract Session: Biomarkers and Adverse Events
Friday, October 9, 2015
Room: Poster Hall
Posters
  • IDWeek_poster_JUS_printetOCT15_a.pdf (671.4 kB)
  • Background:  The incidence and clinical importance of acute liver impairment in critically ill patients is debated. Hyaluronic acid (HA) is released from connective tissue, hepatically eliminated and has been shown to reflect liver function and prognosis in other populations. The aim of the current study was: i) to determine how frequent intensive care patients suffer from clinically relevant liver impairment,  ii) to determine whether this potential liver impairment has impact on 28-day survival. 

    Methods:  A 1,200 intensive care patient cohort from a randomized trial (>80% infected). Patients were excluded if stored serum specimens were too sparse for HA analysis, and if the liver biomarkers bilirubin, INR and Model for End-stage Liver Disease (MELD) were not available at baseline.  All patients with chronic liver failure were excluded. In the final cohort of 839 patients (fig 1) all had HA measured in an immunoturbidimetric assay (HA in healthy controls is ~40 ng/mL).  

    Results:  Biomarker levels at baseline were (median [IQR]), HA (ng/mL):  153.6 [67.9 – 459.5], INR: 1.3 [1.1 – 1.6], bilirubin (mg/dL): 0.53 [0.29 – 0.88], MELD: 13.9 [9.9 – 20.4]. Significant correlations were present between all liver markers (rho 0.15 – 0.56). The mortality risk corresponded directly to the HA quartile they belonged to. The 28-day mortality was ~55% for patients in the 4th quartile vs. ~22% in 1st quartile, logrank, p<0.0001 (fig. 2).  In a multivariable Cox regression model adjusted for known and suspected predictors of mortality, HA quartile III (HR 1.5 [95% CI: 1.0 – 2.5]) and IV (HR 1.9 [95% CI: 1.3 – 2.9]) were found to be strong independent predictors of mortality in intensive care patients (ref Q1). Substantially higher risk was also found in the upper quartile for bilirubin (HR 1.6 [95%CI: 1.1 – 2.3]).

    Conclusion:  Liver impairment, measured by three liver biomarkers and MELD, was frequent in these critically ill patients and was highly predictive for mortality. It is biologically plausible that patients who suffer from liver impairment in critical illness will also be less likely to be able to react adequately to a severe infection and therefore at a higher risk of dying.  The mechanisms causing this liver impairment should be explored to detect targets for interventions to improve outcome.      

    Jens Ulrik Jensen, MD, PhD1, Lars Peters, MD2, Maria Egede Johansen, MD, PhD3, Theis Skovsgaard Itenov, MD4, Morten Bestle, MD, PhD4, Anne Øberg Lauritsen, MD5, Thomas Mohr, MD, PhD6, Katrin Thormar, MD7, Jesper Løken, MD8, Peter Søe-Jensen, MD9, Per Hjort Christensen, MSc10, Mads Holmen Andersen, MD11, Bettina Lundgren, MD, DMSc12, Jesper Grarup, DVM13 and Jens Lundgren, MD, D.M.Sc, Professor1, (1)Chip, Department of Infectious Diseases, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark, (2)Chip, Department of Infectious Diseases and Rheumatology, Rigshospitalet, Copenhagen, Denmark, (3)Chip/Department of Infectious Diseases and Rheumatology, Rigshospitalet, Copenhagen, Denmark, (4)Anesthesia and Intensive Care, North Zealand Hospital, Hillerød, Denmark, (5)Anesthesia and Intensive Care, Glostrup Hospital, Glostrup, Denmark, (6)Anesthesia and Intensive Care, Gentofte Hospital, Hellerup, Denmark, (7)Anesthesia and Intensive Care, Bispebjerg Hospital, Copenhagen NW, Denmark, (8)Anesthesia and Intensive Care, Hvidovre Hospital, Hvidovre, Denmark, (9)Anesthesia and Intensive Care, Herlev Hospital, Herlev, Denmark, (10)Diagnostics, Dako diagnostics, Glostrup, Denmark, (11)Anesthesia and Intensive Care, Aarhus University Hospital, Aarhus, Denmark, (12)Diagnostic Centre, Rigshospitalet, Copenhagen East, Denmark, (13)Department of Infectious Diseases and Rheumatology, Chip, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark

    Disclosures:

    J. U. Jensen, None

    L. Peters, Corgenix: Grant Investigator , Research support

    M. E. Johansen, None

    T. S. Itenov, None

    M. Bestle, None

    A. Lauritsen, None

    T. Mohr, None

    K. Thormar, None

    J. Løken, None

    P. Søe-Jensen, None

    P. H. Christensen, None

    M. H. Andersen, None

    B. Lundgren, None

    J. Grarup, Corgenix: Grant Investigator , Research grant

    J. Lundgren, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.