BACKGROUND: Blood pressure is an important determinant of renal perfusion and acute kidney injury (AKI) is common in hospital patients. However, there is limited knowledge concerning the incidence of relative hypotension prior to its development in general wards. METHODS: We compared blood pressure recordings in a cohort of consecutive patients with no change in renal function to a cohort of patients with acute changes in renal function according to RIFLE classes R, I and F for AKI. We assessed blood pressure over a 3-day period before the development of AKI in index patients and a similar 3-day period in controls. We excluded patients with absolute hypotension [systolic blood pressure (SBP) <90 mmHg]. RESULTS: Patients were old (mean age 76.1 +/- 15.1) and mostly female (57.1%). Those who developed AKI had a lower diastolic blood pressure (P = 0.01), a trend towards lower mean arterial pressure (P = 0.077) and a greater decrease in mean systolic (P < 0.0001), mean diastolic (P < 0.0001) and mean arterial pressure (P < 0.0001) compared to controls. On multivariate logistic regression analysis, a decrease in SBP relative to pre-morbid value was a significant independent predictor of the development of AKI and of RIFLE classes I and F (odds ratio 1.084 for every -1 mmHg change in SBP). CONCLUSIONS: Relative hypotension is more common in ward patients who develop nosocomial AKI than in controls. In these patients, a decrease in SBP relative to pre-morbid value is a significant independent predictor of the development of severe AKI.
BACKGROUND: Blood pressure is an important determinant of renal perfusion and acute kidney injury (AKI) is common in hospital patients. However, there is limited knowledge concerning the incidence of relative hypotension prior to its development in general wards. METHODS: We compared blood pressure recordings in a cohort of consecutive patients with no change in renal function to a cohort of patients with acute changes in renal function according to RIFLE classes R, I and F for AKI. We assessed blood pressure over a 3-day period before the development of AKI in index patients and a similar 3-day period in controls. We excluded patients with absolute hypotension [systolic blood pressure (SBP) <90 mmHg]. RESULTS:Patients were old (mean age 76.1 +/- 15.1) and mostly female (57.1%). Those who developed AKI had a lower diastolic blood pressure (P = 0.01), a trend towards lower mean arterial pressure (P = 0.077) and a greater decrease in mean systolic (P < 0.0001), mean diastolic (P < 0.0001) and mean arterial pressure (P < 0.0001) compared to controls. On multivariate logistic regression analysis, a decrease in SBP relative to pre-morbid value was a significant independent predictor of the development of AKI and of RIFLE classes I and F (odds ratio 1.084 for every -1 mmHg change in SBP). CONCLUSIONS: Relative hypotension is more common in ward patients who develop nosocomial AKI than in controls. In these patients, a decrease in SBP relative to pre-morbid value is a significant independent predictor of the development of severe AKI.