Targeted Exercise Intervention to Reduce Morbidity and Mortality in Sepsis

Participants will perform an entry passive exercise test within 48 hours of onset of severe sepsis. Patients then will perform 30-60 min supine passive cycle ergometry exercise 5 days/week. Details of frequency and duration of each training session will be recorded along with measures of heart rate, hemoglobin oxygen saturation and blood pressure.

The exercise tests will consist of baseline, 30 minutes of supine passive leg cycle ergometry, and a subsequent 60 minute recovery period. Arterial and venous blood samples obtained during the three protocol phases will be analyzed for metabolic factors (blood gases, oxygen content, pH, lactate, glucose). On day 0, 1, and 7, an additional blood tube is immediately transferred on ice to the Translational Research Centre Biohazard Level 2 Laboratory for processing and aliquoting using a standard operating procedure. Plasma will be stored at -80C and thawed immediately before use in biochemical analyses or in vitro experiments.

Analysis will include inflammatory mediator levels, blood vessel permeability in leg muscle and leg muscle perfusion as described below. Additional measures of central and peripheral hemodynamics will include ultrasound-based measures of cardiac stroke volume, carotid blood flow, and carotid artery compliance. Continuous analog measures of the electrocardiogram, central venous pressure, blood pressure, end tidal carbon dioxide and electromyography (quadriceps) will be monitored during the exercise and stored online for subsequent analysis of heart and respiratory rate variability, blood pressure variability and baroreflex sensitivity. We will also record length of ICU stay, in-hospital mortality and hospital health care costs. Length of ICU and hospital stay will be used as surrogates for hospital costs. Physiotherapy time will also be recorded for each patient (time on the intervention and total time with each enrolled patient). Energy requirements will be assessed by metabolic measurements; nitrogen balance and serum prealbumin will be measured to monitor nutritional status.

The mechanistic basis of the observed hemodynamics during and following passive leg exercise will be explored using a comprehensive translational model that incorporates 1) venous blood analysis, 2) in vitro human-derived cell culture models, and 3) in vivo skeletal muscle measures of capillary blood flow. This panel will be evaluated on day 0, 7 and 28.

• Venous Blood Analysis: Patient inflammatory profiles will be determined by assaying stored plasma samples for inflammatory biomarkers and inflammatory-inducing actions. We will measure a previously established list of 20+ inflammatory biomarkers elevated in sepsis, including cytokines, chemokines, matrix metalloproteinases, collagen breakdown products, neurotrophins and stress hormones (cortisol). The innate immune response will also be evaluated. HIF1alpha, a subunit of the transcription factor HIF1, is a key regulator of the innate immune response to inflammation. Exercise results in increased free DNA in the plasma which leads to induced HIF1alpha expression. Thus, HIF1alpha is a potential candidate to orchestrate gene expression changes that initiate adaptations of the innate immune response after exercise. HIF-1alpha, downstream target genes and plasma free DNA levels in blood samples will be measured to determine the effects of exercise on the innate immune response during sepsis.
• In vitro cell culture: Human plasma will be applied to human-derived endothelial cells representing various vascular beds (e.g. cerebral and dermal microvasculature, and large blood vessels such as aorta and veins) to assess inflammatory potential: measurable targets include reactive oxygen species, nitric oxide, NF-kB, MAP kinases, leukocyte adhesion assays, and permeability. Plasma will also be applied to erythrocytes to determine inflammatory-induced changes in morphology, an assay critical to understanding oxygen carrying capacity.
• In vivo capillary blood flow: The occurrence of muscle hypoperfusion and increased vascular permeability will be assessed using an in-house-developed near-infrared spectroscopy (NIRS) method that monitors the uptake of a light-absorbing dye in the muscle. The dye, indocyanine green (ICG), has been used clinically for over 30 years and has a high safety record. Muscle perfusion and vascular permeability are determined by using a mathematical model to characterize the dynamics of the ICG distribution in muscle following an intravenous injection. This approach is analogous to contrast-enhanced imaging methods (CT and MRI), but with the advantage that the portability of NIRS enables bedside measurements.

Inclusion criteria

• Severe sepsis defined as:

• 2 of 4 Systemic Inflammatory Response Syndrome criteria,
• a confirmed or clinically suspected source of infection and,
• two or more organ dysfunction (using standard definitions).

Exclusion criteria

• Greater than 48h since severe sepsis criteria were met
• inability to perform passive exercise using bedside leg cycle,
• limitations on life support interventions (No CPR or No defibrillation by themselves are not an exclusion),
• major limitation in neurologic or physical function prior to onset of severe sepsis (e.g. wheelchair or nursing home dependent),
• patients receiving neuromuscular blocking drugs or sedation targeted to a MAAS score less than 3,
• allergies to sulfonamide drugs or iodides due to possibility of an allergic reaction to ICG.