Included patients
We conducted a prospective study in an 18-bed intensive care unit (ICU) in a tertiary teaching hospital. During a 6-month period, patients older than 18 years admitted for septic shock were included. Septic shock was defined according to the Third International Consensus Definitions for Sepsis and Septic Shock [17]. We included resuscitated patients within the first 24 hours of vasopressor initiation. Exclusion criteria were pregnancy, forearm skin lesions, important soft tissue edema and agitation.
After initial resuscitation, including antibiotic administration, fluid infusion (30 mL/Kg) and norepinephrine infusion to maintain a MAP>65 mmHg, as well as infection focus control when available, patients received intravenous (IV) vitamin C (40 mg/Kg) over 30 minutes. We compared global hemodynamic and tissue perfusion parameters before and 1 hour after vitamin C supplementation, as well as skin microvascular endothelial reactivity (see below). Vasopressor dose was not modified during microvascular investigations.
Assessment of skin microcirculation reactivity
The skin microvascular reactivity was measured in the forearm area by transdermal iontophoresis of acetylcholine (Ach) [18]. This non-invasive technique allows local transfer of Ach across the skin, which produces vasomotor action on subcutaneous capillaries [19, 20]. Ach solution and a weak electrical current are applied onto the skin, creating local differences in electrical potential and active migration of ions and molecules bearing a net electrical charge through epithelial layers. The direction and speed of migration can be adjusted using polarity and the current’s magnitude. The total amount of Ach delivered into the skin is related to the current and duration of application (ie. electrical charge). Acetylcholine acts as an endothelium-dependent vasodilator [21], which induces the production of nitric oxide (NO) after stimulation of the endothelial NO-Synthase. Next, NO induces smooth muscle cells relaxation by activating guanylate cyclase that is responsible for vasodilation and increased blood flow.
The iontophoresis drug delivery chamber was attached to the flexor surface of the non-dominant forearm. The negative lead of the current source was attached to the drug delivery chamber, and the positive lead (ie., reference electrode) to a conductive hydrogel pad fixed onto the wrist. After measurement of baseline blood flow for 60 seconds, three successive applications of Ach were made, every 60 seconds, using anodal current (0.12 mA for 12 seconds each). The drug delivery chamber was loaded with 80 μL of Ach (Miochol®). Variations of blood flow in the skin were assessed by Laser-Doppler Flowmetry technique. A Laser-Doppler Flowmeter probe (Periflux 5000, Perimed), embedded within a heating drug delivery chamber, was used in combination with a current-controlled delivery device (PeriIont, Perimed). Laser-Doppler Flowmeter signals were recorded continuously using an interfaced computer with acquisition software (Perisoft, Perimed).
Baseline blood flow (BF) and area-under-the-curve (AUC) of BF recorded during a standardized 10-minute period were recorded (Additional file 1). Skin blood flow monitoring and analysis were performed by an independent physician who did not participate in patient care.
Data collection
Patients’ characteristics were prospectively collected: age, sex, previous chronic illness, severity of illness evaluated by the Sequential Organ Failure Assessment score (SOFA score) at inclusion [22], source of sepsis, mode of mechanical ventilation, and vasopressor dose. Biological parameters, global hemodynamic parameters [mean arterial pressure (MAP), heart rate (HR)] and cardiac output measured using echocardiography were recorded at 2 time points. In addition, several tissue perfusion parameters were collected at baseline and 1 hour after Vitamin C supplementation: arterial lactate level, index and knee capillary refill time and mottling score, skin temperature and central-skin temperature gradient.
Plasma levels of Vitamin C
Vitamin C plasma levels were measured by a high performance liquid chromatography (HPLC) method adapted from Speek et al. [23]. Briefly, heparinized plasma is stabilized by diluting samples (1/10; v/v) with 5% (w/v) metaphosphoric acid solution. The samples remain frozen at -80°C until assayed. After alcalinization of samples with sodium acetate 4,5 mM (respectively 1 mL and 0,2 mL) and action of ascorbate oxydase (25 µL solution 62 U/mL in Na H2PO4 4mM pH 5,6; 5 minutes at 37°C), the total vitamin C of the sample is converted into acid L-dehydroascorbate. This compound is derivatized with ortho-phenylenediamine (300µL OPDA 100 mM-water solution, 30 minutes 37°C) giving a fluorescent quinoxaline. The vitamin C assay is performed by HPLC in reverse phase with fluorimetric detection. The column is an Intersil C18 ODS2 5µM 4,6x 150 mm. The mobile phase (H2PO4 50mM / methanol (500/214 ; v/v) pH 7,4) flow is 1,15 mL / minute and the injection volume 20 µL. The excitation is done at 346 nm and emission at 424 nm. All the reagent are from Sigma-Aldrich, the column is from Interchim, the HPLC system is a Summit Dionex-Thermo.
Statistics
Continuous variables were presented as median and 25th–75th interquartile ranges (IQRs). Discrete variables were presented as percentages. Comparisons before and after vitamin C injection were made with a paired non-parametric test. Statistical analysis and graphical representations were performed using GraphPad Prism 8.4.1 software (Graph Pad Software Inc., La Jolla, CA). A two-sided p value of less than 0.05 was considered statistically significant.
Ethics
The protocol was approved by an institution’s ethical committee -Comité de Protection des Personnes (CPP Ile de France France, 2019-A03199-48). All patients or their families gave their consent for the study (ClinicalTrials.gov Identifier: NCT04778605).