Cardioventilatory Coupling in Critically Ill Patients

Intrathoracic pressure oscillations due to positive pressure ventilation induce cyclic modifications of activity of the pulmonary vascular receptors and cardiac mechanoceptors. These effects changes the autonomic nervous system modulation with modification of both sympathetic and vagal activity.

Cardio-ventilatory interaction recently has been studied through the heart rate variability in critically ill mechanical ventilated patients. The analysis of the power spectrum density provides information about the power (the variance of beat-to-beat interval) is distributed as frequencies' function. Healthy spontaneously breathing subjects show a cyclic inspiratory increase and expiratory decrease of heart rate, and 'phase coupling' between heart beats and respiration (causing heart beats to occur at constant phases of the respiratory cycle), commonly known as cardioventilatory coupling. These phenomena originate from a complex interplay of several mechanisms including central drive, feedback from arterial baroreceptors, feedback from thoracic and lung stretch receptors, and non-neural mechanisms intrinsic to the heart, which are not fully understood. Although the physiological importance of respiratory sinus arrhythmia and cardioventilatory coupling have not been elucidated, several authors suggested that they might improve ventilation/perfusion matching through a redistribution of heart beats (and consequently of perfusion) within the respiratory cycle, with beneficial effects on gas exchange. Furthermore, decreased respiratory sinus arrhythmia amplitude has been used as an indicator of impaired autonomic control and of poor clinical outcome, also during mechanical ventilation.It has been shown that during controlled mechanical ventilation the respiratory sinus arrhythmia amplitude is considerably reduced and the cardioventilatory coupling generally abolished. Theoretically, mechanical ventilation modes that assist the respiratory pump upon triggering by the patient, such as pressure support ventilation (PSV) might maintain higher respiratory sinus arrhythmia levels through centrally-originated phasic vagal modulation compared to controlled mechanical ventilation. In addition, mechanical ventilation with breath-by-breath variable tidal volumes, so-called 'variable ventilation', could better preserve respiratory sinus arrhythmia and cardioventilatory coupling, and this might play a role in the improved arterial oxygenation found in different models of acute lung injury, although with variable results, when comparing variable and conventional mechanical ventilation. In anesthetized experimental animals the cardioventilatory coupling was more preserved during pressure assisted ventilation than pressure controlled ventilation. If these findings could be present in the humans is yet unknown. Furthermore, it has been demonstrated that some new methodologies of assisted ventilation, such as the neurally adjusted ventilatory assist (NAVA), increase the breath-to-breath variability and tidal volume variability, but their effects on cardioventilatory coupling are not understood.

The study aims to measure the heart rate variability, the respiratory rate variability and the cardioventilatory coupling in critically ill patients during mechanical ventilation both in controlled mode (pressure controlled) and in assisted mode (pressure support ventilation and NAVA).

Methods enrolled patients are connected to a S/5 ICU monitor (GE, Helsinki, Finland) and are mechanically ventilated with a Servo-I ventilator (Maquet, Germany) provided with diaphragmatic electrical activity module (EAdi, Maquet, Germany). A nasogastric 16-Fr EAdi catheter is positioned in all patients. A sequence of three consecutive study phases of different mechanical ventilation modes (PCV, PSV, and NAVA) is started. The sequence of ventilatory modes is randomized for every patient. After a 10 min acclimation period for each study phase, electrocardiographic, arterial pressure and ventilatory waves are collected for consecutive 30 min to a laptop pc via S/5 Collect (GE, Helsinki, Finland) and NAVA Tracker (Maquet, Germany) software for Windows.

Heart rate variability analysis Linear analysis Sequences of 300 consecutives heart beats are selected inside each experimental phase. The mean and the variance of heart period are expressed in msec and msec^2 respectively. Autoregressive spectral density is factorized into components each of them characterized by a central frequency. A spectral component is labeled as LF if its central frequency is between 0.04 and 0.15 Hz, while it is classified as HF if its central frequency is between 0.15 and 0.5 Hz. The HF power is considered to represent respiration-driven vagal modulation of heart rate. To rule out the effect of changes of total power spectrum densities on LF and HF components, spectral values are also expressed in normalized units (NU). Normalization consisted in dividing the power of a given spectral component by the total power minus the power below 0.04 Hz (Very Low Frequency [VLF] spectral component), and multiplying the ratio by 100. The ratio of the LF power to the HF (LF/HF) is considered an indicator of the balance between sympathetic and vagal modulation directed to the heart.