Associations of Fatty Liver Disease with Recovery After Traumatic Injury

Katherine Stern | 2022

Advisor: Ali Rowhani-Rahbar

Research Area(s): Injury & Violence

Full Text

Fatty liver disease (FLD) is prevalent among US adults (20-30%) and associated with chronic systemic inflammation, extra-hepatic comorbidities, and socioeconomic risk factors for poor health outcomes. These factors may directly and indirectly contribute to an individual’s recovery from traumatic injury. However, FLD has not been studied as a risk factor for poor outcomes in the trauma population. We performed a retrospective cohort study of adults admitted to a level-1 trauma center in Seattle, Washington between September 2016 and December 2020, excluding those with severe head injury or cirrhosis. To classify FLD, we measured the liver-spleen attenuation difference in Hounsfield Units (HUL-S) using virtual non-contrast reconstructions of CT scans performed using dual-energy: none (HUL-S>1), mild (-10≤HUL-S<1), or moderate/severe (HUL-S<-10). The primary outcomes of interest were 1) recovery from a systemic inflammatory response (SIRS), defined as SIRS <2, and 2) recovery from organ dysfunction, defined as a sequential organ failure assessment score <2. For each outcome, patients with scores <2 sustained for at least 3 consecutive days or at the time of being discharged alive were classified as recovered. We used Cox models to compare the “hazard” of recovery among those with FLD compared to the hazard of recovery among those without FLD. The secondary outcomes were hospital acquired infection, sepsis, ICU days, and discharge to a skilled nursing or long-term care facility. Relative risks (or rates) for the associations between FLD and outcomes were assessed using multivariable log-binomial and Poisson models. In an exploratory analysis, we compared nested Cox models using the likelihood ratio test to determine whether associations of FLD with the physiologic recovery outcomes differed among patients with higher vs. lower numbers of pre-existing comorbidities or injury characteristics. All models adjusted for age, sex, alcohol use disorder, and health insurance. Among 510 adults meeting inclusion criteria, the median age was 51 years (interquartile range 30, 69), 350 (69%) male, median injury severity score 17 (10, 24). FLD was present in 80 (16%) individuals; 51 had mild FLD and 29 had moderate/severe FLD. Among those with, compared to those without FLD, the adjusted hazard ratio of recovery from SIRS was 0.94 (95% CI 0.72, 1.23), and that for organ dysfunction was 1.08 (95% CI 0.84, 1.41). Associations of FLD with infection, sepsis, ICU days, and disposition were not statistically significant. The association of FLD with recovery from SIRS differed according to whether an individual had shock on admission (HR 0.76 (95% CI 0.55, 1.05) among those with shock, HR 1.81 (95% CI 1.43, 2.28) among those without shock, p-value for interaction = 0.039). Fatty liver disease is common in adults hospitalized after injury. In this study, which was limited by small sample size, FLD was not associated with adverse short-term outcomes. Associations of FLD during recovery from critical illness and long-term outcomes after trauma remain unclear and warrant further study.