Aims/hypothesisThe aim of this study was to assess the risk of death during hospital admission for diabetic ketoacidosis (DKA) and, subsequently, following discharge. In addition, we aimed to characterise the risk factors for multiple presentations with DKA.MethodsWe conducted a retrospective cohort study of all DKA admissions between 2007 and 2012 at a university teaching hospital. All patients with type 1 diabetes who were admitted with DKA (628 admissions of 298 individuals) were identified by discharge coding. Clinical, biochemical and mortality data were obtained from electronic patient records and national databases. Follow-up continued until the end of 2014.ResultsCompared with patients with a single DKA admission, those with recurrent DKA (more than five episodes) were diagnosed with diabetes at an earlier age (median 14 [interquartile range 9–23] vs 24 [16–34] years, p < 0.001), had higher levels of social deprivation (p = 0.005) and higher HbA1c values (103 [89–108] vs 79 [66–96] mmol/mol; 11.6% [10.3–12.0%] vs 9.4% [8.2–10.9%], p < 0.001), and tended to be younger (25 [22–36] vs 31 [23–42] years, p = 0.079). Antidepressant use was greater in those with recurrent DKA compared with those with a single episode (47.5% vs 12.6%, p = 0.001). The inpatient DKA mortality rate was no greater than 0.16%. A single episode of DKA was associated with a 5.2% risk of death (4.1 [2.8–6.0] years of follow-up) compared with 23.4% in those with recurrent DKA admissions (2.4 [2.0–3.8] years of follow-up) (HR 6.18, p = 0.001).Conclusions/interpretationRecurrent DKA is associated with substantial mortality, particularly among young, socially disadvantaged adults with very high HbA1c levels.Electronic supplementary materialThe online version of this article (doi:10.1007/s00125-016-4034-0) contains peer-reviewed but unedited supplementary material, which is available to authorised users.

The safety of carvedilol and other non-selective beta-blocker drugs in patients with liver cirrhosis and ascites is still debated. In this study, we have shown that carvedilol therapy in these patients was associated with reduced risk of mortality, particularly in those with mild ascites. We concluded that low dose, chronic treatment with carvedilol in patients with liver cirrhosis and ascites is not detrimental.

Our data indicate that ROS formation, rather than cellular steatosis per se, impairs mitochondrial function. Thus, reduction in cellular steatosis may not always be the desired outcome without concomitant improvement in mitochondrial function and/or reducing of ROS formation.