Pinsino, A.

Introduction: Sarcopenia, characterized by loss of muscle mass and function, is prevalent in heart failure (HF) and associated with poor outcomes. We investigated alterations in sarcopenia index (SI), a surrogate marker of skeletal muscle mass, in HF, left ventricular assist device (LVAD) and heart transplant (HT) and assessed its relationship with inflammation and digestive tract (gut and oral) microbiota. Hypothesis: Sarcopenia will be associated with increased inflammation, dysbiotic microbial signatures, higher HF class, and earlier time post-HT or post LVAD. Methods: We enrolled 460 HF, LVAD and HT patients. Repeated measures pre and post procedures were obtained in a subset of LVAD and HT patients. Sarcopenia index (serum Creatinine/Cystatin C) and inflammatory biomarkers (C-reactive protein, interleukin-6, tumor necrosis factor-alpha) were measured in 271 and 622 blood samples, respectively. Gut and saliva microbiota were assessed via 16S rRNA sequencing among 335 stool and 341 saliva samples. Multivariable regression models were used to assess the relationship between SI and i) New York Heart Association class; ii) pre- vs. post-LVAD or HT; iii) biomarkers of inflammation and microbial diversity. Results: Median (interquartile range) ln-SI was -0.13(-0.32,0.05). Ln-SI decreased across worsening HF class, further declined within the 1-month after LVAD and HT and rebounded over time to the levels of symptomatic HF. Ln-SI demonstrated an inverse correlation with inflammation (r=-0.28, p<0.0001), and a positive correlation with gut (r=0.26, p<0.0001) and oral microbial diversity (r=0.24, p<0.0001). Presence of the gut taxa Roseburia inulinivorans was associated with increased SI. Conclusions: SI levels decreased in symptomatic HF and remained decreased long-term after LVAD and HT. SI levels covaried with inflammation, gut and oral microbiota in a similar fashion.

Introduction: Serum creatinine (sCr) is routinely used to calculate estimated glomerular filtration rate (eGFR) in heart failure (HF) pts. However, changes in muscle mass may limit the accuracy of sCr as marker of renal function in this population. Cystatin C (CysC) is independent of muscle mass and provides an alternative measure of eGFR. In prior studies, higher sCr/CysC ratio has been associated with higher muscle mass. Herein, we compared CysC- and sCr-based eGFR at serial time points among pts admitted with HF. We hypothesized that muscle mass would decline during HF admission and this would result in a decrease of sCr/CysC ratio. Methods: We pooled pts from 3 trials performed in pts admitted with HF (DOSE, ROSE and CARRESS-HF). eGFR was calculated using CKD-EPI-CysC equation (eGFRCysC) and sCr-based MDRD equation (eGFRsCr). The relative difference between eGFRCysC and eGFRsCr (ΔeGFR) was calculated as follows: (eGFRCysC-eGFRsCr)/ eGFRsCr. To control for confounders, we analyzed changes in sCr/CysC ratio and in ΔeGFR among pts with samples available at both admission and a subsequent time point. Results: A total of 2849 samples were available in 841 pts (age 68 ± 13, 26% female, left ventricular ejection fraction 36 ± 17%). Compared with eGFRsCr, eGFRCysC reclassified 50% of the samples to different GFR categories, mainly to more advanced renal dysfunction (Fig. A). eGFRCysC was significantly lower than eGFRsCr at all time points (Fig. B). From time of admission to all subsequent time points, sCr/CysC ratio declined while ΔeGFR widened (all p-values<0.001). At time of enrollment in CARRESS-HF, each additional day of HF admission was associated with a decline in sCr/CysC ratio of 1.5% (p=0.04). Conclusions: The use of CysC reclassifies a large proportion of pts admitted with HF to more advanced renal dysfunction when compared to sCr-based assessment. The discrepancy between CysC- and sCr-based eGFR appears to widen during HF admission, likely due to muscle mass wasting.