What Is Phrenic Nerve Stimulation and How Does It Work?

Phrenic nerve stimulation is a medical technique that uses electrical impulses to activate the phrenic nerves, which are the nerves that tell your diaphragm to contract and pull air into your lungs. When disease or injury disrupts the brain’s automatic breathing signals, a device can step in and send those signals artificially, restoring something close to natural breathing. The technology has been used for decades in people with high spinal cord injuries, and more recently it has gained traction as an implantable treatment for central sleep apnea and as a way to keep the diaphragm from wasting away in ICU patients on ventilators.

How the Phrenic Nerve Controls Breathing

The phrenic nerve is the sole motor nerve responsible for making the diaphragm move. It originates from a cluster of nerve cells in the mid-cervical spinal cord (roughly the C3 to C5 vertebrae), travels down through the chest, and branches across the diaphragm muscle. Every breath you take starts with a signal traveling down these nerves. The phrenic neuromuscular system does more than just keep you breathing, though; it also plays a role in coughing, swallowing, speaking, and even maintaining posture.¹ When the brain’s respiratory centers fail to send signals, or when the spinal cord is damaged above where the phrenic nerve originates, the diaphragm goes still. You can breathe only with mechanical ventilation, or with a device that stimulates the phrenic nerve directly.

How Stimulation Devices Deliver the Signal

There are several ways to get an electrical pulse to the phrenic nerve, and the method depends on the medical condition being treated.

Surgically implanted diaphragm pacers involve placing electrodes directly on or near the phrenic nerve through a procedure in the chest or neck. These systems have been used since the 1960s in patients with spinal cord injuries and in children born with conditions that impair their automatic breathing drive. The electrodes connect to a small receiver implanted under the skin, which picks up signals from an external transmitter worn on the body. The patient or caregiver can adjust settings like breathing rate and the strength of each pulse.

Transvenous phrenic nerve stimulation takes a less invasive approach. Instead of opening the chest, a doctor threads a thin lead through a vein and positions it near where the phrenic nerve runs alongside certain blood vessels. The remedē System, for instance, places a stimulation lead in the left pericardiophrenic vein or the right brachiocephalic vein, with a separate sensing lead in the azygous vein to detect the patient’s breathing pattern.² The whole setup resembles a cardiac pacemaker implant, with the pulse generator tucked into the pectoral region. Studies comparing transvenous and direct phrenic nerve stimulation have found a strong correlation between the diaphragm pressures generated by each method, confirming that the transvenous approach can effectively replace direct stimulation for evaluating and driving diaphragm function.³

A third approach, percutaneous stimulation, uses needle-like electrodes placed temporarily through the skin near the phrenic nerve in the neck. This is primarily used in ICU settings where the goal is short-term diaphragm conditioning rather than permanent breathing support.

Treating Central Sleep Apnea

Central sleep apnea is different from the more common obstructive variety. In obstructive sleep apnea, the airway physically collapses. In central sleep apnea, the brain intermittently stops sending the signal to breathe during sleep. This is common in people with heart failure and can be difficult to treat, since standard CPAP therapy addresses a mechanical problem that isn’t the core issue here.

Transvenous phrenic nerve stimulation has become an FDA-approved option for moderate-to-severe central sleep apnea. The implanted device detects when the patient’s breathing drops off during sleep and delivers gentle pulses to the phrenic nerve to trigger a diaphragm contraction, essentially filling in the gaps that the brain leaves. In the pivotal trial, treatment produced a roughly 55% reduction in apnea-hypopnea index (a measure of how many times breathing is disrupted per hour) within three months.⁴ At six months, about 60% of treated patients had achieved at least a 50% drop in their breathing-disruption score, compared with only 11% in the untreated control group.⁵

Those benefits appear to hold up over time. Five-year follow-up data showed that breathing disruptions remained significantly reduced, with the median number of events per hour dropping from 46 at baseline to 17 at five years. Central apneas specifically fell from a median of 23 per hour to just 1, though some residual shallow breathing events persisted. Daytime sleepiness scores also improved meaningfully.⁶ Sleep quality improved beyond just fewer apneas: arousal events dropped substantially, time spent in REM sleep increased, and the percentage of sleep spent with low oxygen levels was cut roughly in half.⁵ When former control-group patients had their devices activated after the initial study period, they saw the same magnitude of improvement as the original treatment group, confirming that the benefits were reproducible and not just a placebo effect.⁷ The improvements held regardless of whether patients had previously tried positive airway pressure therapy.⁸

Some patients do require additional steps during the implant procedure. In about a quarter of cases in one series, balloon angioplasty or another vascular technique was needed to clear a path for the lead to reach the right spot near the phrenic nerve.⁹ The serious adverse event rate across 12 months was about 9%, which is comparable to other implantable cardiac-rhythm devices.⁷

Breathing Independence After Spinal Cord Injury

High cervical spinal cord injuries, at the C1 to C4 level, can sever the connection between the brain’s respiratory center and the phrenic nerves. The diaphragm itself and the phrenic nerves below the injury site may be perfectly healthy, but they never receive the signal to contract. These patients typically require a mechanical ventilator 24 hours a day. Phrenic nerve pacing can restore their ability to breathe without being tethered to a machine.¹⁰

Before surgery, doctors need to verify that the phrenic nerve and diaphragm below the injury are still functional. This is done with phrenic nerve conduction studies, where the nerve is stimulated externally while fluoroscopy (real-time X-ray) watches for diaphragm movement. If the diaphragm contracts when the nerve is zapped, the patient is a candidate for a permanent implant.¹¹

Results from a European multi-center study found that among 21 patients with at least a year of follow-up, about 38% achieved complete weaning from mechanical ventilation, using the pacer around the clock. The median daily use across all patients was 15 hours. Complications were uncommon, limited to one device revision and one wire issue.¹² Case reports from other settings have documented ventilator-free breathing sustained for years after implantation, with one report describing a tetraplegic patient breathing independently at 36 months post-implant with no respiratory infections or procedure-related complications.¹³

Conditioning the diaphragm is a critical part of the process. A diaphragm that has been idle for months or years while on a ventilator is weak and fatigues quickly. Stimulation starts at low intensity and short durations, with gradual increases over weeks to months. In children with high spinal cord injuries, researchers have achieved continuous ventilatory support through simultaneous stimulation of both sides of the diaphragm, with some patients sustaining this for years without signs of fatigue.¹⁴

Children Born With Impaired Breathing Drive

Congenital central hypoventilation syndrome (sometimes called Ondine’s curse) is a rare genetic condition in which the brainstem fails to properly control automatic breathing, especially during sleep. Children with this condition may stop breathing whenever they fall asleep and require lifelong ventilatory support. Diaphragm pacing offers these children a way to move freely during the day and, for some, to eventually have their tracheostomy tube removed.

Among 18 children treated with diaphragm pacing in one series, about 72% were successfully ventilated by the pacer without a tracheostomy. Of those who had a tracheostomy before the pacer was implanted, 73% were eventually decannulated.¹⁵ This is a meaningful shift in quality of life: a child who no longer needs a tracheostomy and ventilator tubing can participate more freely in school, play, and social activities.¹⁶ Diaphragm pacers also give full-time ventilator-dependent patients greater mobility, since the external transmitter is small and portable compared with a ventilator unit.¹⁷

Protecting the Diaphragm in the ICU

A newer application of phrenic nerve stimulation targets a problem that happens inside intensive care units: ventilator-induced diaphragm dysfunction. When a patient is placed on a mechanical ventilator, the machine does the work of breathing and the diaphragm sits idle. Within days, the diaphragm starts to atrophy, much like a leg muscle wasting away in a cast. This weakness makes it harder to wean patients off the ventilator, prolonging ICU stays and raising the risk of complications.

Animal studies have shown that stimulating the diaphragm during mechanical ventilation preserves the muscle’s strength and prevents the atrophy that otherwise develops. In one study, neurostimulation on every breath preserved lung oxygenation and reduced the loss of end-expiratory lung volume after 50 hours of ventilation. The stimulated animals also had lower airway pressures, better lung compliance, and more uniform lung inflation.¹⁸ The idea is straightforward: if you can keep the diaphragm exercising even while the ventilator is doing most of the breathing work, the muscle stays in shape for when the patient needs to breathe on their own again.

In human ICU patients, early feasibility studies have demonstrated that phrenic nerve stimulation is safe and can increase diaphragm thickness and excursion.¹⁹ Larger randomized trials are now underway. One pivotal trial is testing a percutaneous system designed to temporarily stimulate both phrenic nerves in patients who are having difficulty weaning from the ventilator, with the goal of shortening the time to independent breathing.²⁰ If these trials succeed, phrenic nerve stimulation could become a routine part of ICU care for ventilated patients.

The ALS Controversy

Amyotrophic lateral sclerosis (ALS) is a degenerative disease that progressively destroys motor neurons, including those that control breathing. Because phrenic nerve stimulation had worked well in patients whose nerves were intact but disconnected from the brain (like spinal cord injury), there was early enthusiasm for trying it in ALS, where the motor neurons themselves are dying. The logic was that exercising the diaphragm might slow its decline.

An initial pilot study reported encouraging signs: the diaphragm appeared to move better under stimulation, muscle thickness increased, and the decline in lung function seemed to slow. Median survival from implant was about 20 months.²¹ But a subsequent randomized controlled trial told a starkly different story. The DiPALS trial compared non-invasive ventilation alone against non-invasive ventilation plus diaphragm pacing, and it had to be stopped early on safety grounds. Patients who received diaphragm pacing survived a median of 11 months, compared with 22.5 months in the control group. The pacing group also had more than twice the rate of adverse events.²² The conclusion was unambiguous: diaphragm pacing should not be used as a routine treatment for ALS patients in respiratory failure.²³

The likely explanation is that in ALS, the motor neurons supplying the diaphragm are already damaged and dying. Forcing them to fire repeatedly through electrical stimulation may accelerate their degeneration rather than strengthening the muscle. This stands in sharp contrast to spinal cord injury, where the neurons and muscle are healthy and simply lack the brain’s “go” signal. The ALS experience is an important reminder that phrenic nerve stimulation depends on the underlying nerve and muscle being viable. When the hardware is broken rather than disconnected, stimulation can do harm.

Unexpected Benefits Beyond Breathing

Patients with spinal cord injuries who switch from mechanical ventilation to diaphragm pacing often report improvements that go well beyond breathing itself. In one study, patients who had been on positive-pressure ventilation scored in the anosmia or severe microsmia range on a standard smell test. After switching to diaphragm pacing, their scores jumped to normal or near-normal. The reason: positive-pressure ventilation pushes air into the lungs, which means air does not flow naturally through the nasal passages. Diaphragm pacing restores the negative-pressure breathing cycle, pulling air in through the nose and re-engaging the sense of smell.²⁴

When asked what mattered most, patients ranked improved mobility first, followed by better self-image and relationships, with restored smell and a greater feeling of security tied for third. The mobility improvement makes sense: a small external transmitter is far less cumbersome than a ventilator with hoses and a battery backup. Patients can leave the house, ride in a car, and interact with others without the social barrier of a large, noisy machine.

Closed-Loop and Adaptive Systems

Most current phrenic nerve stimulators operate on fixed or semi-fixed settings. A clinician programs the pulse frequency, amplitude, and breathing rate, and the device delivers those parameters until the next clinic visit. This works well enough in stable conditions, but the body’s breathing needs change constantly with sleep stages, physical effort, and even posture.

Researchers are developing closed-loop systems that can sense the body’s respiratory state and adjust stimulation in real time. One approach uses sensors that measure the airflow produced by each stimulated breath, then automatically adjusts the electrode position and stimulation strength based on the result.²⁵ Another line of work focuses on computational models that simulate gas exchange in the lungs and use those signals as biofeedback to the pacing device, allowing the system to adapt to different metabolic demands and disease states.²⁶ The transvenous system used for central sleep apnea already incorporates a rudimentary version of this idea, with its sensing lead detecting respiratory effort and timing the stimulation to fill in pauses. But fully adaptive pacing that responds to oxygen levels, carbon dioxide, and moment-to-moment ventilatory needs is still in the research phase.

What Phrenic Nerve Stimulation Costs

The upfront expense of a phrenic nerve stimulator is substantial, particularly for surgically implanted systems. The device itself, the surgical procedure, and the initial hospital stay add up to considerably more than starting a patient on a standard mechanical ventilator. But ventilators are not cheap to run over time: they require continuous nursing support, disposable supplies like suction catheters and filters, frequent maintenance, and around-the-clock electricity.

A cost analysis comparing the two approaches in spinal cord injury patients found that the reduced need for airway nursing equipment made phrenic nerve stimulation less expensive on an ongoing basis. The higher initial cost of the stimulator was repaid within about four years.²⁷ For patients expected to need ventilatory support for decades, as is the case with many young spinal cord injury patients and children with congenital hypoventilation syndrome, that payback period is a small fraction of the device’s useful life. The harder-to-quantify savings come from reduced hospital readmissions for ventilator-related infections and the social and economic value of greater independence.