Cyclic changes in arterial pulse during respiratory support revisited by Doppler echocardiography.

It has long been known that there are cyclic changes in arterial pressure during mechanical ventilation. They are caused by cyclic changes in both the right and left ventricular stroke output, occurring in opposite phases. As a result, arterial pulse pressure is increased during inspiration and decreased during expiration. A cyclic improvement in left ventricular systolic function could thus be expected during mechanical lung inflation. We tested this hypothesis in 31 septic patients who were mechanically ventilated in controlled mode by combining left ventricular measurements by transesophageal echocardiography with invasive arterial pressure recordings and Doppler analysis of pulmonary venous flow and right and left ventricular stroke volume. Lung inflation by tidal ventilation significantly improved left ventricular stroke volume (26 +/- 0.4 cm3/m2 [mean +/- SEM] vs. 22.3 +/- 0.4 cm3/m2 at end deflation). Beat-to-beat analysis of pulmonary venous flow velocity illustrated the boosting effect of lung inflation on pulmonary venous return. The beneficial effect of inspiration thus appeared directly related to a significant increase in left ventricular diastolic volume (60.3 +/- 1.5 cm3/m2 vs. 53.3 +/- 1.4 cm3/m2 at end-expiration) and to a lesser extent to an improved left ventricular ejection fraction. We concluded that the transient beneficial hemodynamic effect of tidal ventilation on the left ventricular pump is essentially mediated by an improved left ventricular filling.

Increasing respiratory rate to improve CO2 clearance during mechanical ventilation is not a panacea in acute respiratory failure.

BACKGROUND: Increasing respiratory rate has recently been proposed to improve CO2 clearance in patients with acute respiratory failure who are receiving mechanical ventilation. However, the efficacy of this strategy may be limited by deadspace ventilation, and it might induce adverse hemodynamic effects related to dynamic hyperinflation. SETTING: An intensive care unit of a university hospital. PATIENTS: We studied 14 patients with acute respiratory failure during the adjustment of ventilator settings on the first day of mechanical ventilation in volume-controlled mode. MEASUREMENTS: After determining the positive end-expiratory pressure that suppresses any intrinsic positive end-expiratory pressure at a respiratory rate of 15 breaths/min, we compared blood gas analysis, respiratory measurements, and Doppler evaluation of right ventricular systolic function by using two different respiratory strategies with the same airway pressure limitation (plateau pressure, < or =25 cm H2O), a low-rate conventional respiratory strategy with a respiratory rate of 15 breaths/min, and a high-rate strategy with a respiratory rate of 30 breaths/min. RESULTS: Compared with the low-rate strategy, the high-rate strategy neither significantly reduced PaCO2 (47 +/- 8 vs. 51 +/- 7 mm Hg with the low-rate strategy) nor significantly improved PaO2 (99 +/- 40 vs. 95 +/- 35 mm Hg with the low-rate strategy). It significantly increased alveolar deadspace to tidal volume ratio (21% +/- 8%, vs. 14% +/- 6% with the low-rate strategy) and produced dynamic hyperinflation, resulting in a substantial intrinsic positive end-expiratory pressure (6.4 +/- 2.7 cm H2O). Right ventricular outflow impedance was increased, resulting in a significant drop in the cardiac index (2.9 +/- 0.6 vs. 3.3 +/- 0.7 L/min/m with the low-rate strategy). CONCLUSION: We conclude that a high respiratory rate strategy during mechanical ventilation in patients with acute respiratory failure did not improve CO2 clearance, produced dynamic hyperinflation, and impaired right ventricular ejection.

Pressure-volume curves in acute respiratory distress syndrome: clinical demonstration of the influence of expiratory flow limitation on the initial slope.

The presence of an initial segment with a low compliance on the static pressure-volume (PV) curve in patients with acute respiratory distress syndrome (ARDS) indicates that some lung compartments do not initially receive insufflated gas. We tested the hypothesis that an uneven distribution of time constants, producing a "slow compartment," was in part responsible for the change in compliance between the initial and the intermediate segment of the PV curve. In 16 patients with ARDS submitted to mechanical ventilation in volume-controlled mode with a supportive respiratory rate of 15 breaths/minute, we constructed the static PV curve on the first day of respiratory support and determined the intrinsic positive end-expiratory pressure (PEEPi4) during a prolonged end-expiratory pause (4 seconds). We also measured the volume of a "slow compartment" during a prolonged expiration (> 6 seconds), and determined an external PEEP (PEEPe) suppressing PEEPi4. Among the 16 patients studied, 11 exhibited a low inflection point, associated with a "slow compartment" of 172 +/- 83 ml, responsible for a PEEPi4 of 3 +/- 2 cm H2O. Conversely, the five remaining patients had a linear PV curve, associated with a minimal "slow compartment" of 28 +/- 10 ml, responsible for a negligible PEEPi4. We observed that individual slopes of the initial segment of the PV curve were inversely and significantly correlated with the proportion of the "slow compartment" (r = -0.85). We concluded that the shape of the inspiratory PV curve in ARDS might be dependent on the presence of a "slow compartment," and demonstrated that a low external PEEP appeared sufficient to achieve a substantial mechanical improvement in clinical practice.