Ethics

This study was approved by the Ethics Committee of Agence Régionale de Santé Occitanie (FRANCE, president D. Benayoun) (NOR-PHARM: 1-17-42 Clinical Trials: NCT03454204) (05/03/2018). This prospective observational study was conducted in the Department of Anaesthesiology and Critical Care of the University Hospital of Lariboisière, Paris, France. All methods were carried out in accordance with relevant guidelines and regulations. Written informed consent was obtained from all subjects, and/or their legal guardian(s), participating in the trial.

Patients

The inclusion criteria were as follows: adult patients (> 18 and older) undergoing major surgery (mostly neurosurgical interventions). Only patients for whom preoperative anaesthesia consultation indicated continuous ABP and CO monitoring were screened for inclusion.

The exclusion criteria were as follows: patients aged under 18 years old, pregnancy, haemodynamic instability caused by acute bleeding, critical haemodynamic instability, sitting position and preoperative doses of norepinephrine > 500 µg/hour.

Patients’ characteristics and management

All cardiovascular, respiratory, and neurologic comorbidities, ASA (American Society of Anaesthesiologists) scores, usual medications, clinical abnormalities, and results of preoperative diagnostic tests were recorded on the anaesthesiology report. Patients were treated in accordance with local care protocols.

All general anaesthesia were conducted under intravenous administration of propofol and remifentanil guided by AIVOC mode (with Schneider model for propofol and Minto model for remifentanil). The bispectral index guided the deepness of anaesthesia with objectives between 40 and 60. Patients’ lungs were mechanically ventilated in volume control mode (tidal volume: 6 ml/kg) with an O2/air mixture (FiO2 0.4) and a positive end expiratory pressure of 5 cmH2O. The respiratory rate was adjusted to keep end-tidal carbon dioxide partial pressure (EtCO2) between 35 and 40 mmHg.

Standard monitoring included HR, pulsed oximetry (SpO2), perfusion index (PI) [23], invasive ABP, temperature, and ventilatory settings, including end tidal CO2 (EtCO2). CO and SV monitoring was performed via oesophageal Doppler (Deltex Medical Limited, Chichester, UK).

All variables, including CO and stroke volume (SV) were recorded continuously and extracted via the Xtrend Logiciel from Philips® (Amsterdam, Netherlands) scopes.

Cardiac afterload was monitored via the Global After Load Angle (GALA). Its value was visualized on the transoesophageal Doppler screen and calculated from the velocity pressure loops (VP loops) as previously described [24].

Microcirculation monitoring

Microcirculation monitoring was performed simultaneously with 5 different devices. All the devices and the main scope monitor were connected and synchronized, allowing to find retrospectively all the desired events as the peak of MAP.

Videocapillaroscopy using incident dark field imaging device (Braedius®, Netherlands) : intraoperative sublingual video allowed direct visualization of sublingual blood flow in microvascular networks. Images acquisition and analysis were performed according to international guidelines [25]. Care was provided to avoid pressure artefacts and saliva was removed as much as possible before acquisition.

If a bolus was administered, all the sequences were recorded in the same zone one after the other. First sequence was considered as the basal state of patient’s microcirculation. When the peak of MAP was reached, the sequence corresponding was kept apart to be analyzed after. This is in line with the new guidelines [25].

In case a continuous infusion was administered, 5 sequences were recorded in 5 different zones at basal state and 5 supplementary ones were recorded when the arterial blood pressure was stabilized at the next level (plateau). Sequences duration was between 8 and 10 seconds.

Two trained operators (MK, FV) recorded the clips.

Analysis was made by four trained investigators (MK, FV, JM, JJ), independently, and blinded from other’s investigator’s results. They were assisted by the last version of the software CytoCamTools Study Manager provided by Braedius® only concerning the total vessel density score. Vessels were added or removed manually if the detection proposed by the software was not satisfactory.

Clips were not randomized before analysis, but analysis was made blinded from the identity of the patient, the intervention (bolus or continuous infusion), and the level of arterial blood pressure at the time of the recording.

Four microcirculatory parameters have been analyzed:

1)

The microvascular flow index (MFI), which is a qualitative evaluation of the microvascular flow. The image is divided into four quadrants, and the predominant type of flow in very small vessels is assessed in each quadrant using an ordinal score (0 = no flow, 1 = intermittent flow, 2 = sluggish flow, 3 = normal flow). The overall score, called microvascular flow index, is the sum of each quadrant score divided by the number of quadrants.

2)

The total vessel density (TVD)

3)

The perfused vessel density (PVD), which is calculated using the formula: PVD = PPVxTVD.

4)

The Heterogeneity Index (HI), calculated as the difference between extreme values of MFI between the five recordings divided by its mean value for the continuous infusion recordings; and as the difference between the best and the worst flow divided on the average flow for bolus recordings.

Laser Doppler (Transonic System®, Ithaca, USA) : This technique is based on the reflection of a beam of laser light that undergoes a change in wavelength (Doppler shift) when moving blood cells are encountered. The magnitude and frequency distribution of these changes are related to the number and velocity of red cells. The quantitative value derived from these measurements (red blood cell flow) is expressed in arbitrary perfusion units (Tissular Perfusion Unit, TPU).

Near-Infrared Spectroscopy technology (NIRS), providing tissular saturation in O2 (StO2) (InSpectra®, Hutchinson Technology, Minnesota, USA) : two sensors placed on the patients’ thenar eminence emit near-infrared light in four specific wavelengths (690, 780, 805 and 850 nm), which is then partially absorbed and partially reflected to the sensors, based on the ratio of oxygenated to deoxygenated hemoglobin in the tissue. NIRS also measures THI (tissular hemoglobin index), an index of microcirculatory volume of hemoglobin, with values ranging from 0 to 100.

Tissular CO2 measure (TOSCA 500, Radiometer®,Copenhague, Danemark) : Transcutaneous measurement of PCO2 is based on the phenomenon of CO2 gas diffusing very easily throughout the body tissue and skin, which can thus be detected by a sensor on the skin surface. tPCO2 reflects tissue perfusion by an evaluation of the difference between tPCO2 and artPCO2, so-called « CO2gap » [26].

Perfusion Index (PI) by photoplethysmography (Two devices : Phillips® and Masimo®)

Examination of microcirculation with an incident dark field imaging, laser Doppler, NIRS technology and transcutaneous CO2 captor is completely noninvasive and respects the noninterventional study design.

Laser Doppler and tissular CO2 probes were each placed on one ear as previously described in other publications [23]. The NIRS probe was placed on the thenar eminence [27], contralateral to the arterial catheter. Photoplethysmography by Philips® was placed on one index contralateral to the arterial catheter.

Protocolized management of IOH

The mean arterial blood pressure (MAP) reference was the average between MAP at anaesthesia consultation and MAP the day before surgery. IOH episode was defined as a decrease in MAP under 65 mmHg or below 20% of the MAP reference in the absence of clear hypovolemia. Hypovolemia was defined in our institution by the conjunction of a significant decrease in CO compared to the CO target value obtained after CO optimization before surgical incision, a DeltaPP >15% and/or an acute bleeding episode.

When a patient presented an IOH episode, the anaesthesiologist in charge could:

Either administered a bolus of 2 mL (10 µg) of norepinephrine diluted at 5 µg/mL.

Or either introduce a continuous infusion of the same treatment at speed 40 mL/hour (i.e., 200 µg/hour or 16 µg/5 min). If a continuous infusion was already started, an elevation of 40 ml/hour more was added.

The protocol was not randomized. As explained, the decision of a bolus or a continuous infusion was not made by the investigators of the study, but by the anaesthesiologist in charge of all the intervention. Often but not systematically, continuous infusion is given if the arterial pressure continues to be low despite some one or two first boluses. Patients could receive two boluses but it had to be with an interval of more than 10 minutes. Patients receiving a bolus directly followed by a continuous infusion were not included.

Collection of macro- and microcirculatory parameters over time:

For the bolus, the analysis times were T0, T15, T30, T45, T60 seconds and at peak of MAP. This choice derives from a study we are conducting (nonpublished data) which shows that the peak in MAP is reached at approximately 50 seconds. For videocapillaroscopy sequence analysis, the times were T0 and at the peak of ABP increase.

For the continuous infusion, the analysis times were T0, T180, T300 seconds and at peak of MAP. For videocapillaroscopy, times were T0 and at the ABP plateau (plateau being defined by less than 5% variations in PAM for more than 20 seconds).

Values of each parameter at peak of effect and time to reach this peak were also recorded.

Outcomes

The primary outcome MFI measured by videocapillaroscopy. Secondary outcomes were CO, MAP, SV, TVD, PPV, HI, PI, TPU, StO2 and tPCO2.

Statistics

Statistical analysis was performed with GraphPad Prism® software (La Jolla, California, USA). The significance level was α= 0,05. Data with a normal distribution are presented as the mean ± standard deviation, and those with a nonnormal distribution are presented as the median ± interquartile range. The normality of the population distribution was determined using a Shapiro test. Variations in the parameters over time were always compared with the baseline value using a Wilcoxon test and expressed as a percentage. To compare the values obtained with a bolus and with a continuous infusion at each time point including the peak of effect (peak of MAP), we performed a Mann–Whitney test. As the number of boluses and continuous infusion varied from one patient to another, we used weighted methods to compare those average values, with the weights being the number of measures per subject.

Inter- and intraindividual variability for videocapillaroscopy analysis was assessed by calculating the kappa coefficient of concordance.