Background. Previous studies have reported that spontaneous oscillations at low frequencies are diminished at the onset of septic shock in adult patients, which has been suggested to reflect dysfunction of the cardiovascular autonomic nuclei (1,2). We hypothesised that a suppression of spontaneous blood pressure oscillations to lower frequencies also occurs, and furthermore evaluated whether changes in these oscillations are adversely associated with outcome.Methods. We retrospectively collected invasive steady state blood pressure recordings from 64 patients admitted to the intensive care unit with severe sepsis or septic shock. We used the Kaplan-Meier log-rank estimator to assess univariate differences in 30-day mortality between the lower- and upper two third tertiles of low-frequency oscillations, and a Cox-based time-to-event regression to develop a multivariable survival model.Results. We found that spontaneous low and mid-frequency blood pressure oscillations occurred at markedly lower frequencies than those reported from healthy humans, at 0.009 (0.007-0.010) Hz and 0.02 (0.01-0.03) Hz, respectively. Less pronounced or absent low frequency blood pressure oscillations was associated with a higher 30-day mortality (HR 3.6 [95% CI: 1.4-8.9, p=0.0062]) (figure).Conclusion. Mid- and low frequency blood pressure oscillations operate at much reduced frequencies in patients with severe sepsis or septic shock; furthermore, the absence of spontaneous low frequency oscillations may herald an adverse outcome. Our findings support the notion that cardiovascular failure in sepsis may involve failure of the autonomic system, conceivably due to dysfunction of cardiovascular autonomic nuclei in the brainstem.
Physiology 2014 (London, UK) (2014) Proc Physiol Soc 31, PCB114
Poster Communications: Evidence for a suppression of spontaneous blood pressure oscillations in mechanically ventilated patients with sepsis
R. M. Berg1,4, R. R. Plovsing2, A. M. Greve3, C. B. Christiansen1, L. N. Toksvang1, N. Holstein-Rathlou4, K. Møller5,1
1. Centre of Inflammation and Metabolism, Rigshospitalet, Copenhagen, Denmark. 2. Intensive Care Unit 4131, Rigshospitalet, Copenhagen, Denmark. 3. Department of Cardiology, Rigshospitalet, Copenagen, Denmark. 4. Renal and Vascular Research Section, University of Copenhagen, Copenhagen, Denmark. 5. Neurointensive Care Unit 2093, Rigshospitalet, Copenhagen, Denmark.
View other abstracts by:
Where applicable, experiments conform with Society ethical requirements.