Red blood cell distribution width (RDW) is known to be a predictor of severe morbidity and mortality in some chronic diseases such as congestive heart failure. However, to our knowledge, little is known about RDW as a predictor of mortality in patients with Gram-negative bacteremia, a major nosocomial cause of intra-abdominal infections, urinary tract infections, and primary bacteremia. Therefore, we investigated whether RDW is an independent predictor of mortality in patients with Gram-negative bacteremia. Clinical characteristics, laboratory parameters, and outcomes of 161 patients with Gram-negative bacteremia from November 2010 to March 2011 diagnosed at Severance Hospital, Yonsei University College of Medicine, Seoul, Korea, were retrospectively analyzed. The main outcome measure was 28-day all-cause mortality. The 28-day mortality rate was significantly higher in the increased RDW group compared with the normal RDW group (P < 0.001). According to multivariate Cox proportional hazard analysis, RDW levels at the onset of bacteremia (per 1% increase, P = 0.036), the Charlson index (per 1-point increase, P < 0.001), and the Sequential Organ Failure Assessment score (per 1-point increase, P = 0.001) were independent risk factors for 28-day mortality. Moreover, the nonsurvivor group had significantly higher RDW levels 72 h after the onset of bacteremia than did the survivor group (P = 0.001). In addition, the area under the curve of RDW at the onset of bacteremia, the 72-h RDW, and the Sequential Organ Failure Assessment score for 28-day mortality were 0.764 (P = 0.001), 0.802 (P < 0.001), and 0.703 (P = 0.008), respectively. Red blood cell distribution width at the onset of bacteremia was an independent predictor of mortality in patients with Gram-negative bacteremia. Also, 72-h RDW could be a predictor for all-cause mortality in patients with Gram-negative bacteremia.

No abstract

No abstract

Little is known about the cellular characteristics of CD8(+) T cells in rheumatoid arthritis (RA). We addressed this by investigating whether the frequency of the CD8(+) T cell subsets and their phenotypic characteristics are altered in the peripheral blood and synovial fluid (SF) from patients with RA. In this study, CD8(+) T cells, mainly CD45RA(-) effector memory (EM) CD8(+) T cells, were increased significantly in the SF, but not in the peripheral blood from RA patients, compared with healthy controls. The synovial EM CD8(+) T cells were activated phenotypes with high levels of CD80, CD86, and PD-1, and had a proliferating signature in vivo upon Ki-67 staining, whereas the Fas-positive cells were prone to apoptosis. In addition, EM CD8(+) T cells in the SF were less cytotoxic, as they expressed less perforin and granzyme B. In particular, the proportions of synovial fluid mononuclear cells that were CCR4(+)CD8(+) T cells and IL-4-producing CD8(+) T cells (i.e., Tc2 cells) were significantly higher than those in peripheral blood mononuclear cells of patients with RA and healthy controls. In addition, the number of IL-10-producing CD8(+) suppressor T (Ts) cells increased significantly in the SF of RA patients. Especially, CD8(+) T cells were inversely correlated with disease activity. These findings strongly suggest that EM CD8(+) T cells in the SF are increased, likely because of inflammation, and they may be involved in modulating inflammation, thereby affecting the development and progression of RA.

The p21-activated kinase, Shk1, is required for cell viability, establishment and maintenance of cell polarity, and proper mating response in the fission yeast, Schizosaccharomyces pombe. Previous genetic studies suggested that a presumptive protein methyltransferase, Skb1, functions as a positive modulator of Shk1. However, unlike Shk1, Skb1 is not required for viability or mating of S. pombe cells and contributes only modestly to the regulation of cell morphology under normal growth conditions. Here we demonstrate that Skb1 plays a more significant role in regulating cell growth and polarity under conditions of hyperosmotic stress. We provide evidence that the inability of skb1⌬ cells to properly maintain cell polarity in hyperosmotic conditions results from inefficient subcellular targeting of F-actin. We show that Skb1 localizes to cell ends, sites of septation, and nuclei of S. pombe cells. Hyperosmotic shock results in substantial delocalization of Skb1 from cell ends and nuclei, as well as stimulation of Skb1 protein methyltransferase activity. Taken together, our results demonstrate a new role for Skb1 as a mediator of hyperosmotic stress response in fission yeast. We show that the protein methyltransferase activity of the human Skb1 homolog, Skb1Hs, is also stimulated by hyperosmotic stress in fission yeast, providing evidence for evolutionary conservation of a role for Skb1-related proteins as mediators of hyperosmotic stress response, as well as mechanisms involved in regulating this novel class of protein methyltransferases.