Prediction of mortality in patients with sepsis using detrended fluctuation analysis of Heart Rate Variability

Sepsis is one of the common causes of mortality in critical care units. In the present study, we compared the last 25 hours of heart rate dynamics of survived and non-survived patients with sepsis admitted in ICU. We calculated the RR-interval time series of patients and detrended fluctuation analysis (DFA) was performed on each 30 minute periods of data sets. The difference of scaling exponent (α) between the two groups was significant from 9 hours before the death. In deceased patients, the scaling exponent of RR-intervals decreased along the time approaching the death from values close to 1 (in healthy states) toward 0.5, indicating a qualitative change in the fractal-like structure of the heart rate series which makes it different from a physiologically relevant 1/f dynamics.