A comprehensive description of kidney disease progression after acute kidney injury from a prospective, parallel-group cohort study.

Acute kidney injury (AKI) is associated with adverse long-term outcomes, but many studies are retrospective, focused on specific patient groups or lack adequate comparators. The ARID (AKI Risk in Derby) Study was a five-year prospective parallel-group cohort study to examine this. Hospitalized cohorts with and without exposure to AKI were matched 1:1 for age, baseline kidney function, and diabetes. Estimated glomerular filtration rate (eGFR) and the urinary albumin:creatinine ratio (uACR) were measured at three-months, one-, three- and five-years. Outcomes included kidney disease progression, heart failure episodes and mortality. In 866 matched individuals, kidney disease progression at five years was found to be significantly increased in 30% of the exposed group versus 7% of those non-exposed (adjusted odds ratio 2.49 [95% confidence interval 1.43 to 4.36]). In the AKI group, this was largely characterized by incomplete recovery of kidney function by three months. Further episodes of AKI during follow-up were significantly more common in the exposed group (odds ratio 2.71 [1.94 to 3.77]) and had an additive effect on risk of kidney disease progression. Mortality and heart failure episodes were more frequent in the exposed group, but the association with AKI was no longer significant when models were adjusted for three-month eGFR and uACR. In a general hospitalized population, kidney disease progression after five years was common and strongly associated with AKI. Thus, the time course of changes and the attenuation of associations with adverse outcomes after adjustment for three-month eGFR and uACR suggest non-recovery of kidney function is an important assessment in post-AKI care and a potential future target for intervention. STUDY REGISTRATION: ISRCTN25405995.

Structure and variation in end-of-life discussions in the Surgical Intensive Care Unit.

The research reported here is an exploratory discourse analysis of a corpus of six end-of-life discussions in a Surgical Intensive Care Unit (SICU), describing the structure and variations of the four phases of an end-of-life discussion in terms of the function of each of these phases: the Opening (Phase 1), Description of Current Status (Phase 2), Holistic Decision Making (Phase 3), and Logistics of Dying (Phase 4). Of particular interest is Phase 2, in which the presentation of medical information culminates in an inferential summary statement that functions to establish the patient's status as terminal. We argue that it is Phase 2 that is crucial in the functional progression of an end-of-life discussion toward a decision to move from therapeutic to palliative care, since it is in Phase 2 that physicians and families interactionally achieve a consensus that allows a decision to withdraw or withhold further treatment, including life support, which would be futile and only prolong the patient's suffering. We show how two of the end-of-life discussions in the corpus that did not establish the terminal status of the patient in Phase 2 did not move to decision making in Phase 3.