NAVA in the Neurologically Injured

Significant advances in mechanical ventilation are rarely the product of paradigm- changing technology; the blending of modes, the creation of sensitive triggers, and the permission of different amounts of patient interaction coupled with varying support have defined the majority of recent advances.

However, the introduction of neurally adjusted ventilatory assist (NAVA) has been a step in a radically different direction for the provision of mechanical ventilatory support. NAVA is a near-universally applicable mode of ventilation that can be used in adults and children, including neonates, and is unaffected by weight or cardiopulmonary pathology.

This mode of ventilation is initiated and controlled by the neural signals causing actual depolarization of the diaphragm. The direct relationship to the diaphragmatic signal allows for synchronous ventilation in the face of spontaneous efforts and will provide an assist pressure that is proportional to the electrical signal, thereby coupling the direct output of the respiratory centers of the brainstem with the physiologically driven respiratory effort and desired effort intensity of the patient.1

The measurement of the intensity of signal conduction down the phrenic nerve to the diaphragm is measured by means of a specialized nasogastric tube that is inserted while using a catheter placement screen on the ventilator while at the bedside. (See Figure 1.) The depolarization signal is measured in microvolts (µV) and is directly proportional to the force at which the neural respiratory centers are driving the diaphragm to contract.

Archive Image6

Click to view larger graphic.

However, in the setting of critical illness, the ability of the diaphragm to perform the work that is directed by the phrenic-relayed signals is often greatly impaired. Therefore, rather than relying on the ability of the diaphragmatic muscle fibers to perform the work of ventilation, the signals are intercepted by the nasogastric tube, displayed as an Electrical Diaphragmatic Impulse (EDI), and the mechanically delivered pressure is proportional to the integral of the EDI signal (EAdi). (See Figures 2 and 3.)

The proportionality factor, or the operator-determined support, can be adjusted based on the clinical scenario and the desired ratio of support to patient effort. Increasing the patient’s “NAVA level” typically leads to an increase in delivered pressure for a given EDI, or it may lead to the patient decreasing the intensity of the signal of diaphragmatic depolarization, thereby leading to a decrease in the EAdi produced by the patient and allowing the inspired pressure to remain essentially unchanged. A combination of effects also may be seen. This varies significantly from standard pressure support modes where the pressure is constant and does not vary once the patient has triggered the support breath.

An obvious application of NAVA technology is in the neurologically ill patient, whose decrease in diaphragmatic function and respiratory failure is neither a failure of the phrenic signal nor the inherent strength of the diaphragm, but a failure of signal conduction at the motor end-plate of the diaphragm. We have used NAVA ventilation in the neurosciences intensive care unit (NSICU) at our institution in a number of scenarios.

Case example: myasthenia gravis

A middle-aged Caucasian female with a severe myasthenic crisis was admitted to the NSICU at our institution for impending respiratory failure, quickly requiring intubation and mechanical ventilation. She was noted to have marked dyssynchrony on standard volume-control SIMV and was requiring greater and greater sedation. Despite standard maneuvers and attempts at MV with pressure control, volume control or pressure support, we were unable to establish adequate synchrony and attempted NAVA ventilation.

Upon placement of the NAVA NG tube she was noted to have a measurable EDI signal, and she was placed on NAVA support (0.5cm H20/µV) and immediately reported improved comfort and tolerated rapid weaning of sedation. After one hour with the NAVA support, she had an EDI signal noted between 6-10 µV (normal adult EDI 8-10 µV) and tidal volumes (Vt) of 4-10 ml/kg of ideal body weight and blood gas measurements that demonstrated resolution of the hypercarbic respiratory failure that had initiated the original intubation.

Remarkably, the patient remained in severe crisis with profound motor weakness and no evidence of axonal recovery for a number of days; however, her diaphragm remained partially active, and we were able to synchronize her respiratory efforts based on phrenic signaling allowing less sedation and continued conditioning of her respiratory musculature. She was weaned to extubation without difficulty.

Case example: cervical spine injury

An intubated middle-aged Caucasian male was admitted to the NSICU with a traumatic cervical spine injury. Clinically, he was determined to be a C-2 level, complete, ASIA-A spinal cord injury with quadraparesis and presumed limited respiratory muscle capacity. He received a tracheostomy early in his care. Soon after, a NAVA-capable NG tube was placed, and his EDI waveform was monitored. We noted that he had preserved EDI signal generation, indicating that his diaphragm was receiving signals from his central respiratory centers via his phrenic nerve and providing the possibility for weaning of MV support.

We were able to use the EDI in the context of his vital sign monitoring to assure that he was not in physiologic distress during his attempted weaning. We found that an increasing EDI signal without a paired rise in his Vt indicated that he was beginning to fatigue and additional NAVA support was required. Likewise, if his Vt increased without an increase in his EDI, we were able to continue his MV weaning. By following paired EDI signals and Vt, it was remarkably clear when a patient was fatiguing, which allowed us to prevent over-aggressive weaning yet still manage to wean an C-2 level ASIA-A patient to freedom from mechanical ventilation.

Case example: hydrocephalus

A young adult female with hydrocephalus was admitted to the NSICU for increasing ICP and worsening neurological status, despite cerebrospinal fluid (CSF) diversion. She was noted to be extremely difficult to ventilate as a result of a failure to attain adequate synchrony. The NAVA NG tube was placed, and her EDI signal was evaluated in her current, non-NAVA mode of MV to assess for neural synchrony. Her dyssynchrony was persistent regardless of attempted mode of MV, and she was transitioned to NAVA in an attempt to improve synchrony and provide effective ventilation.

During the period of EDI monitoring and transition to NAVA, we noted erratic EDI signals, both in frequency and amplitude; these signal anomalies coincided with a worsening of her neurologic exam despite an unchanged ICP that was being monitored continuously. After nearly an hour, during which there was no hypercarbia, she developed a sudden dramatic rise in her ICP and was taken emergently to the OR for decompressive hemicraniectomy. Upon returning to the NSICU, her EDI signal normalized along with her neurologic exam and her ICP. She was ultimately discharged home after ventriculoperitoneal (VP) shunt placement.

This case is possibly the first to capture EDI signal variation in the setting of the “pre-herniation” phase of ICP rise when the brain contents are beginning to shift inside the cranial vault despite normal or near normal ICP readings. Remember, ICP readings are often regional and not reflective of the true pressure that the entire brain is being subjected to inside the cranial vault. While more patients will require EDI signal monitoring during this period to confirm our finding, it is entirely possible that an erratic and bizarre EDI signal in the setting of intracranial pathology may be the herald of impending herniation.

Case example: TBI

Patients with severe TBI can be challenging to ventilate. In one such case, we cared for a severe TBI patient in whom we could not achieve synchrony regardless of the mode of ventilation used. We noted that the patient had a combination of missed triggers and variable timing of respiratory effort. At times his abdominal muscles seemed to tense and contract without a clear cause. An astute member of the team suggested that the patient might be suffering dyssynchrony as a result of the rhythmic, regular diaphragmatic contractions seen with singultus (hiccoughs).

A NAVA NG was placed, and we were able to clearly demonstrate that there was no EDI signal during breath triggering and that there were several missed triggers as well. It was then apparent that the patient had abdominal muscle involvement and not diaphragmatic misfiring as a cause of his dyssynchrony. We slowly weaned the patient’s sedation and continued MV with NAVA, successfully extubating the patient later.

Unique method

NAVA ventilation offers a unique method of synchronizing ventilatory support to patient effort. By synchronizing MV with the central nervous signals transmitted along the phrenic nerve, NAVA allows coupling of patient effort to ventilatory support in a sophisticated and physiologically sound manner. Our descriptions of the use of NAVA in the neurologically injured are examples of the ability to expand this MV technology to disease states other than ARDS/ALI and may offer practitioners another powerful tool in the provision of MV to patients with respiratory failure from non-pulmonary causes.


1. Terzi N, Pelieu I, Guittet L, et al. Neurally adjusted ventilatory assist in patients recovering spontaneous breathing after acute respiratory distress syndrome: Physiologic Evaluation. Crit Care Med. 2010;38:1830-37.

Kipp Leihgeber, RRT, NPS, is a respiratory therapist at UC Health-University Hospital, Cincinnati. He has received consulting honoraria from MAQUET Medical. Jordan Bonomo, MD, is assistant professor, emergency medicine director in the division of critical care, department of emergency medicine; assistant professor, neurosurgery/neurocritical care; and co-director, neurocritical care fellowship at the University of Cincinnati. Dr. Bonomo has received grant funding from the NIH, the University of Cincinnati, and from Integra Life Sciences as well as honoraria for consulting from Genentech Inc. and Philips Medical.