[Comparison of three ventilatory modes during immediate postoperative transfer of cardiac surgical patients]

OBJECTIVE

Compare three ventilatory strategies during the immediate postoperative transfer of cardiac surgical patient.

STUDY DESIGN

Prospective, comparative and observational study.

PATIENTS AND METHODS

After approval by our local ethical committee, 330 patients undergoing on-pump cardiac surgery were consecutively included. Patients suffering from chronic obstructive pulmonary disease, exhibiting intraoperative hypoxemia or requiring nitric oxide were excluded. The ventilatory mode was left at the discretion of the anesthesiologist and included: controlled mechanical ventilation (FiO(2)=1, N=124) or (FiO(2)=0.6, N=106), and manual ventilation using rebreathing bag (N=100). A blood gas analysis was performed immediately prior to connecting patient at ventilator at the arrival in ICU.

RESULTS

The mean duration of transfer was 3.9+/-1.4 min. Invasive pressure monitoring was used in all patients. The pulse oxymetry and electrocardiogram were respectively used in 78% and 24% of patients. PaO(2) values less than 100 mmHg and those more than 300 mmHg were more frequently found in patients ventilated by rebreathing bag (42%) and mechanical ventilation FiO(2)1 (52%), respectively. No significant difference was found between groups regarding PaCO(2) values.

CONCLUSION

When rebreathing bag is used for transfer in ICU, severe decrease in PaO(2) may be observed. In absence of intraoperative hypoxemia, a mechanical ventilation with FiO(2)0.6 seems to be the most suitable ventilatory strategy for such short immediate postoperative transfer.

From the grating scale monitor to the generalized seeing monitor

An instrument named the grating scale monitor for measuring the outer scale L0 from the angle-of-arrival (AA) fluctuations of a perturbed wave front was developed a few years ago at Nice University. The AA is detected with a 5-ms time resolution by modulation of the stellar image in a small telescope with a grating. One uses the normalized covariance of AA fluctuations to estimate L0. A new version of this instrument, the generalized seeing monitor (GSM) is described. It consists of four identical modules for measuring the AA at four locations on the wave front. A spatiotemporal analysis of these data leads to the determination of seeing epsilon0, outer scale L0, and the wave-front speed. In addition, isoplanatic angle theta0 is determined from scintillation, making the characterization of turbulence with the GSM almost complete. We describe the instrument and make a detailed analysis of its performance and accuracy. Several site-testing campaigns have been conducted with the GSM: at La Silla (Chile), Oukaïmeden (Morocco), Maidanak (Uzbekistan), and Cerro Pachon and Cerro Paranal (Chile). The main results of these campaigns are presented and discussed.

Analytical solution for the covariance and for the decorrelation time of the angle of arrival of a wave front corrugated by atmospheric turbulence

The two-dimensional (2D) spatial covariance of the angle-of-arrival (AA) fluctuations is often used to investigate the properties of wave fronts corrugated by the atmosphere for high-angular-resolution techniques. Theoretical series expansions of this covariance are presented. The fast convergence of these series reduces the calculation time of the covariance done by numerical integration. The 2D covariance is a nonradial function. A physical interpretation of this anisotropy is proposed. The spatiotemporal correlation of the AA is deduced from the covariance assuming the "frozen-flow" hypothesis. The impact of the anisotropy on the evaluation of the number of predominant turbulent layers and on the corresponding winds is investigated, and an analysis of temporal correlations is performed. A simple theoretical approximation of the decorrelation time of the AA is given, which is found to be in agreement with experimental results.