Objective: To assess the association between glycaemic status prior to the first hospital presentation with developing adverse renal outcomes overtime in patients with multiple hospital re-admissions. Design: A prospective observational cohort study. Participants: All inpatients aged ≥54 years admitted between 2013 and 16 to a tertiary hospital. Main outcomes: We prospectively measured HbA1c levels in all inpatients aged ≥54 years admitted between 2013 and 16. Diabetes was defined as prior documented diagnosis of diabetes and/or HbA1c ≥6.5% (47·5 mmol/L). Included patients had ≥ two admissions (at least 90 days apart), baseline estimated glomerular filtration rate (eGFR) >30 ml/min/1·73m2 and no history of renal replacement therapy. We assessed several renal outcomes: (a) 50% decline in eGFR; (b) rapid decline in renal function (eGFR decline >5 mL/min/1·73m2/year) and (c) final eGFR<30 ml/min/1·73m2. Results: Of 4126 inpatients with a median follow-up of 465 days (254, 740), 26% had diabetes. The presence of diabetes was associated with higher odds of (a) 50% decline in eGFR (OR = 1·42;95% CI:1·18–1·70;p < 0·001); (b) rapid decline in renal function (OR = 1·40;95%CI:1·20–1·63;p < 0·001), and (c) reaching eGFR<30 ml/min/1.73m2 (OR = 1·25;95%CI:1·03–1·53;p < 0·05). Every 1% (11 mmol/L) increase in baseline HbA1c was associated with significantly greater odds of (a) >50% decline in eGFR (OR = 1·07;95% CI:1·01–1·4;p < 0·05) and (b) rapid decline in renal function (OR = 1·11;95% CI:1·05–1·18;p < 0·001). Conclusions: In patients with ≥two admissions, the presence of diabetes and higher HbA1c levels were strongly and independently associated with adverse renal outcomes at follow up. Such patients are at high risk of relatively rapid deterioration in renal function and a logical target for structured preventive interventions.
- Hospitalised patients
- Renal disease