#3305 Progression of renal disease and mortality in patients with CKD who do or do not receive loop diuretics

Abstract Background and Aims Loop diuretics (LD) are often prescribed in patients with chronic kidney disease (CKD) with or without a diagnosis of heart failure, to treat hypertension or symptoms and signs of congestion. The prognostic association of the LD use with change in kidney function and other important clinical outcomes has rarely been evaluated. Method Electronic medical records for adults from the Greater Glasgow region (covering a population of approximately 1.5 million people) who had at least two serum creatinine measurements between 1 January 2012 and 31 December 2014 were analysed. Patients already receiving renal replacement therapy (RRT) were excluded. Estimated glomerular filtration rate (eGFR) was calculated using the CKD-EPI (2009) formula, without adjusting for ethnicity. Patients were stratified according to the last eGFR measured prior to 01/01/2015. LD prescribing immediately prior to 2015 and its association with the eGFR slope, the need for renal replacement therapy, and mortality over the following three years was assessed. Results The cohort comprised 94,006 patients, including 2,152 with eGFR <30, 8,062 with eGFR 30–44, 19,192 with eGFR 45–59, and 64,600 with eGFR >60 mL/min/1/73 m2. LD were prescribed to 12,252 (13%) patients at baseline, with higher rates in patients with lower eGFR (11%, 14%, 21%, and 33.5% for patients with eGFR >60, 45–59, 30–44, and <30 mL/min/1.73 m2, respectively). Amongst those prescribed LD, the prevalence of heart failure was nearly 10 times higher compared to those not receiving LD across all eGFR groups (1.3% vs. 16% in those with eGFR >60 and 3.3% vs. 21.5% in those with eGFR <30 mL/min/1.73 m2) but only a minority of patients treated with LD had a formal diagnostic record of heart failure (∼21% overall). Using a mixed-effects model with individual-level random effects over 3 years, eGFR declined by 2.5 mL/min/1.73 m2/year for the entire cohort. Within 3 years, 474 (0.5%) patients in the entire cohort required RRT. Patients prescribed LD were more likely to receive RRT than those who were not (1.1% versus 0.4%). Mortality at 3 years was considerably higher for those prescribed LD (30%) compared to those who were not (14%), across the spectrum of eGFR (see Table 1). Conclusion Patients with a lower eGFR are more often prescribed LD and have a higher prevalence of heart failure. Prescribing LD is associated with a substantially higher mortality rate even in the absence of a diagnosis of heart failure. This may reflect under-diagnosis of heart failure in this population, an adverse effect of LD on prognosis, or the use of LD may be a pharmaco-epidemiological surrogate marker of undiagnosed heart failure, accelerate cardiovascular disease progression or multi-morbidity.