Publications

AKI occurred in more than 40% of patients after CA. These patients had more severe hemodynamic impairment and needed more aggressive ICU therapy; however the development of AKI did not influence neurological recovery.

Introduction: Although β-lactams have a relatively safe therapeutic index, several cases of neurotoxicity have been reported. We assessed the correlation between β-lactams concentrations and the incidence of electroencephalographic (EEG) abnormalities in septic patients. Methods: We reviewed the data of all septic patients treated in our ICU from January 2012 and July 2013 with meropenem (MEM), piperacillin-tazobactam (PTAZ), aztreonam (AZT) or ceftazidim/cefepime (CEF) and in whom at least one therapeutic drug monitoring (TDM) and one EEG were performed the same day. Trough drug concentrations (Cmin) were measured using high-performance liquid chromatography. Cmin was normalized to the clinical breakpoint of Pseudomonas aeruginosa for each antibiotic (Cmin/MIC) following the EUCAST recommendations. EEG abnormalities included seizure activity, periodic EEG discharges (PEDs) or frontal intermittent rhythmic delta activity (FIRDA). Results: We obtained 181 TDMs (65 MEM, 84 PTAZ, 26 CEF, 6 AZT) in 136 patients. We found 24 (13%) seizure activities, 14 (7%) PEDs and 5 (3%) FIRDA. The proportion of patients with EEG abnormality was significantly correlated with increased Cmin/MIC ratio (p=0.002). This finding remained after exclusion of patients admitted for primary brain injury (p=0.03). EEG abnormalities were correlated with high Cmin/MIC for PTAZ (p=0.01) but not for MEM (p=0.49); CEF and AZT were not evaluated because of the limited number of patients. Conclusions: This is the first study reporting a relationship between high β-lactams concentrations and an increased incidence of EEG alterations in septic patients.

The SOFA-2 score, updated to include contemporary organ support treatments and new score thresholds, describes organ dysfunction in a large, geographically and socioeconomically diverse population of critically ill adults.

The contributions of increases in circulating catecholamines, changes in central command, and muscle afferents on baroreflex control of the sinus node during exercise are unclear. We used a dobutamine infusion to induce hemodynamic changes comparable to those of moderate physical exercise in the absence of changes in central command and muscle afferents in 13 healthy subjects. Dobutamine (up to 9 microg/kg body weight per minute) increased systolic blood pressure, shortened the RR interval, increased systolic blood pressure variability, but blunted RR interval variability (P<0.05 versus placebo). Consequently, dobutamine decreased the coherence between variations in systolic blood pressure and RR interval and decreased arterial baroreflex sensitivity from 12+/-2 to 3+/-1 ms/mm Hg (P<0.01). The largest increases in systolic blood pressure with dobutamine were paralleled by the greatest impairments in arterial baroreflex sensitivity (0. 50<r<0.56, P<0.01). The chronotropic effects of dobutamine prevented a reflex bradycardia in response to the blood pressure increase. However, less predominant low-frequency oscillations in systolic blood pressure (P<0.0001) suggested preserved sympathetic withdrawal in response to the blood pressure increase induced by dobutamine. In conclusion, this study revealed that a shift in the operating point of the arterial baroreceptors and the chronotropic effects of adrenergic stimulation impair baroreflex control of the sinus node during dobutamine exercise stress testing. Baroreflex control of the sinus node is not reset when hemodynamic characteristics of exercise are reproduced in the absence of modifications in central command and muscles afferents.

Introduction: Cells under acute controlled ischemia initiate complex physiological adaptations to be able to better tolerate further hypoxic events. This phenomenon, called ischemic preconditioning, has been well described in myocardial cells, but can also occur in other tissues. We evaluated the potential therapeutic role of ischemic pre- and post-conditioning in experimental sepsis. Methods: We studied 16 sheep (25-35 Kg) monitored invasively and treated with volume controlled mechanical ventilation under anesthesia with midazolam, ketamine and morphine. We induced abdominal sepsis by injecting autologous feces into the peritoneal cavity and followed animals until death or for a maximum of 30 hours. Animals were randomly allocated to undergo ischemic preconditioning followed by intermittent post-conditioning (CONDITIONING) or not (CONTROL). Controlled ischemic episodes of the lower extremities and pelvis were obtained by inflating an intravascular balloon in the aortic bifurcation. We performed 4 cycles of 2 min ischemia (4 min apart) before sepsis induction and thereafter 1 inflation every 4 hours until death. We calculated the time taken to develop diuresis of <0.4 ml/Kg/hour (KIDNEYTIME), a lactate >2 mmol/L (LACTIME), a systemic mean arterial pressure <60 mmHg (HYPOTENSIONTIME), and the time to death (DEATHTIME), all expressed in hours. Time to event analysis was performed using a Log-Rank test with SPSS 19 (IBM, USA). Results: All results are presented as median (p25-p75) comparing CONTROL vs CONDITIONING groups respectively: KIDNEYTIME was 17 (13-20) vs 28 (22-30) p=0.033, LACTIME 18 (15-21) vs 25 (20-30) p=0.101, HYPOTENSIONTIME 17 (14-19) vs 28 (23-30) p<0.001 and DEATHTIME 25 (23-27) vs 30 (27-30) p=0.023. Conclusions: In this sheep model of severe abdominal sepsis, repeated ischemic pre- and post-conditioning of the pelvis and lower extremities resulted in improved hemodynamics, renal function and survival time.