Introduction: There is limited data examining predictors of outcome in patients with HeartMate 3 (HM3) LVAD. We sought to examine factors that predict pump replacement and death in patients with HM3 LVAD. Hypothesis: Traditional risk factors in older LVAD models may be unassociated with pump replacement and death in patients with HM3 LVAD. Methods: Patients that underwent HM3 LVAD implantation from 2015 to 2020 were identified from a single tertiary care center. Records were reviewed for demographic/clinical characteristics at LVAD implantation and a primary outcome of LVAD replacement or death. Predictors of outcome were identified using univariable Cox regression. Kaplan Meier (KM) analysis and log-rank test were used to assess outcome-free survival based on pre-operative BMI and albumin. Results: A total of 298 patients that met inclusion criteria were identified. Over a median follow-up of 2.3 years, 97 patients (33%) experienced a primary outcome. Predictors of outcome included age, peripheral vascular disease, ECMO prior to LVAD implantation, and pre-operative creatinine and albumin levels. BMI was not predictive of the primary outcome (Table 1). Outcome-free survival was different among patients with and without pre-operative hypoalbuminemia, but unassociated with differences in BMI (Figure 1). Conclusions: This is one of the largest single-institution cohorts that has examined predictors of adverse outcomes in patients with HM3 LVAD. Optimizing pre-operative nutrition (as reflected by albumin) may help prevent adverse outcomes in this population. Lastly, HM3 LVAD may be a safe option in obese patients who would otherwise not be candidates for transplantation.

The immune mechanisms for durable immunity against S. aureus skin infections are unclear as reinfections are common despite specific antibodies and Th1/Th17 cells. Using a CA-MRSA skin reinfection model in mice involving a primary infection (1°) in the lower back (cleared by 14 days) followed by a secondary infection (2°) in the upper back on day 28, we found that 1° and 2° wt mice had similar responses. In contrast, 1° IL-1β−/− mice had increased lesions and bacterial burden with impaired neutrophil recruitment whereas 2° IL-1β−/− mice were protected and resembled wt mice. Transfer of serum containing CA-MRSA specific antibodies did not confer protection to 1° IL-1β−/− mice and depletion of CD4+ T cells did not alter the protection of 2° IL-1β−/− mice, indicating a lack of a role of antibodies or CD4+ T cells. However, 2° IL-1β−/− mice were no longer protected if the egress of lymphocytes from lymph nodes was blocked in 2° IL-1β−/− mice (FTY720 treatment) and the immune impairment in naïve IL-1β−/− mice could be rescued by adoptive transfer of total lymph nodes cells from d28 IL-1β−/− mice. Moreover, transfer of only 50,000 γδ T cells from lymph nodes of day 28 IL-1β−/− mice conferred protection to 1° IL-1β−/− mice. These γδ T cells produced IFNγ/TNF-α, but not IL-17/IL-22, and neutralizing IFNγ/TNF-α activity resulted in loss of protection in 2° IL-1β−/− mice. In humans, similar IFNγ/TNF-α-producing γδ T cells were found in individuals with IRAK4 deficiency whose susceptibility to S. aureus skin infections improves with age but not from individuals with chronic granulomatous disease whose susceptibility is lifelong. These findings define an alternative mechanism to therapeutically target for long-term protection against CA-MRSA skin infections.