LAUSR

Objective: Mortality rate is a major healthcare quality indicator and the study of its features is important to improve clinical outcomes. Recent data suggest that blood pressure (BP) variability promotes increased cardiovascular risk independently of BP and may be associated to mortality. The authors evaluated the association between BP variability in acute hospitalized patients in an Internal Medicine ward and its outcome. Design and method: Retrospective study of all patients discharged in the first month of 2015 in an average-size hospital Internal Medicine ward. Odds ratios (OR) and respective 95% confidence intervals (95% CI) were obtained with the use of binary logistic regression to identify independent predictors of hospital mortality, adjusted for confounding variables (age, gender, duration of hospitalization, number of diagnoses, mean systolic BP (SBP) during hospitalization, coefficient variation of SBP (%) (CVSBP) defined as (SD SBP/meanSBP)x100. Analysis of area under the receiver operating characteristic curve (ACU) was performed to determinate the accuracy of the model. Results: In the reported period, 395 patients were studied. The mean age, median number of diagnoses and SBP mean were respectively 80.72 ± 10.99 years, 9 (IQR = 6–13) diagnoses and 127 ± 16 mmHg. 304 patients were discharged and 91 died. In the deceased group the mean age was 84.73 ± 8.87 years, the mean length of hospital stay 12.14 ± 13.93 days, and the median number of diagnosis was 9 (IQR = 6–12). In this group, the mean SBP was 123 ± 17 mmHg and the CVSBP was 14.31 ± 4.38. In the binary logistic regression analysis (ACU 73%), after adjustment of confounding variables, the age and the CVSBP are predictive of mortality during hospitalization, respectively OR 1.07 (95% CI 1.04–1.11 p = 0.000) and OR 1.15 (95% CI 1.07–1.24 p = 0.000). The mean SBP and the number of diagnoses had a negative correlation with mortality, respectively OR 0.98 (95% CI 0.96–0.99 p = 0.008), and OR 0.95 (95% CI 0.91–0.99 p = 0.03). The length of hospital stay (p = 0.54) and gender (p = 0.18) had no statistical significance in the prediction of mortality. Conclusions: These findings emphasize the importance of the SBP variability during hospital stays for acute medical patients, suggesting relevance to understand mortality rate.