HeartMate 3 LVAD: A Heart Pump for Advanced Heart Failure

The HeartMate 3 is a surgically implanted mechanical pump that takes over most of the work of the left ventricle, the heart’s main pumping chamber, in people with advanced heart failure who are no longer responding to medication. It is the only durable left ventricular assist device (LVAD) currently approved in the United States and has largely replaced its predecessors on the strength of a landmark trial showing better survival, fewer strokes, and dramatically lower rates of pump malfunction. For patients facing either a long wait for a heart transplant or no prospect of one at all, the HeartMate 3 has become the standard mechanical option.

How the Pump Works

The HeartMate 3 is a centrifugal-flow pump, meaning blood enters the center of a spinning rotor and is flung outward into the aorta by centrifugal force. What sets it apart from older continuous-flow LVADs is that its rotor is fully magnetically levitated. There are no bearings, no points of mechanical contact, and no seals touching the rotor. The rotor floats inside the pump housing, suspended by magnetic fields and controlled by sensors. This eliminates a major source of friction and heat, both of which contributed to blood clot formation in earlier devices.

The pump also has a built-in artificial pulse feature that rapidly changes rotor speed by several thousand revolutions per minute in a cyclical pattern. This speed modulation was designed to reduce zones of stagnant blood inside the pump. Computational fluid dynamics studies have found that the pump achieves good washout of blood whether or not the artificial pulse is active, which likely explains the very low rates of clots forming inside the device. The artificial pulse does appear to help clear near-wall regions and shift stagnation points, and it creates rapid fluctuations in wall shear stress that may help dislodge early deposits before they can grow into problematic clots.¹

In practical terms, the pump sits in the pericardial space near the heart’s apex. A flexible cable called a driveline exits the body through a small incision in the abdomen and connects to an external controller and batteries that the patient carries. The controller monitors pump function and alerts the patient or clinician to changes in speed, power, or flow.

The MOMENTUM 3 Trial and Survival Data

The evidence base for the HeartMate 3 rests heavily on the MOMENTUM 3 trial, a large randomized study that compared it head-to-head against the older axial-flow HeartMate II. At two years, roughly 77% of patients receiving the HeartMate 3 were alive and free of disabling stroke or pump replacement, compared with about 65% of those receiving the older device. Pump replacement itself was needed far less often: about 2% in the HeartMate 3 group versus 11% with the HeartMate II. Rates of stroke, major bleeding, and gastrointestinal bleeding were all lower with the newer pump.²

The five-year follow-up data were even more striking. The composite of survival to transplant, recovery, or continued LVAD support free of disabling stroke or pump replacement was about 54% in the HeartMate 3 group versus roughly 30% in the axial-flow group. Overall survival at five years was about 58% versus 44%. Serious adverse events including stroke, bleeding, and pump thrombosis remained less frequent with the HeartMate 3 over the full follow-up period.³

Stroke Reduction

Stroke has long been one of the most feared LVAD complications. The MOMENTUM 3 data showed a pronounced difference after the first six months: in the long-term period beyond 180 days, stroke rates in the HeartMate 3 group were more than three times lower than in the HeartMate II group. Statistical analysis identified treatment with the HeartMate 3 as the only independent predictor of fewer strokes.⁴ A separate registry study of destination therapy patients found that having a HeartMate 3 cut the risk of stroke by nearly 60% compared with older continuous-flow pumps, and that postoperative atrial fibrillation was a significant independent risk factor for stroke.⁵

Bridge to Transplant Versus Destination Therapy

LVADs are implanted under two broad strategies. “Bridge to transplant” means the pump is intended to keep the patient alive and functional until a donor heart becomes available. “Destination therapy” means the pump is the final treatment, typically because the patient is ineligible for transplant due to age, other medical conditions, or personal choice. One of the important findings from the MOMENTUM 3 subgroup analysis was that the HeartMate 3 showed a comparable treatment benefit regardless of which strategy was intended at the time of implantation.⁶

That said, outcomes are not identical between the two groups in real-world practice. A single-center retrospective study found that destination therapy patients tended to be older, had more chronic kidney disease, and had lower five-year survival compared with bridge-to-transplant patients. They also experienced more adverse events, particularly kidney problems requiring temporary dialysis and bleeding.⁷ These differences likely reflect the patients’ underlying health profiles more than any limitation of the device itself.

Complications That Persist

While the HeartMate 3 has reduced many of the problems associated with older LVADs, it has not eliminated them. The major complications that patients and clinicians still contend with fall into a few categories.

Bleeding and Acquired Von Willebrand Syndrome

Gastrointestinal bleeding remains the most common serious adverse event after LVAD implantation. A key contributor is acquired von Willebrand syndrome: the mechanical shear forces inside the pump break apart large multimers of von Willebrand factor, a protein essential for normal blood clotting. In most LVAD patients, this breakdown happens rapidly after implantation and impairs the blood’s ability to form plugs at sites of vascular injury.⁸ Studies of axial-flow LVADs found that all patients on the device developed a pattern resembling von Willebrand disease type 2, and that the abnormality reversed after device removal.⁹ A study of continuous-flow LVAD patients showed that gastrointestinal bleeding occurred in about 29% of patients over a median follow-up of roughly 20 months, and that those who bled had significantly lower levels of large von Willebrand factor multimers.¹⁰

The HeartMate 3’s wider blood flow gaps and lower shear stress are expected to reduce this problem compared with older devices, and the MOMENTUM 3 data did show lower bleeding rates. But the phenomenon is inherent to any pump that generates continuous high-velocity flow, so it has not disappeared entirely.

Driveline Infections

The driveline exit site is a permanent opening in the skin and represents a chronic infection risk. These infections can range from superficial skin redness to deep-seated infections involving the pump pocket. Managing them is a long-term challenge that typically involves prolonged courses of antibiotics, sometimes lasting many months, and occasionally surgical debridement or wound vacuum therapy.¹¹ Standardized wound care protocols that emphasize sterile dressing changes and secure driveline immobilization have become central to prevention efforts.¹² Despite these improvements, driveline infection remains one of the most common reasons for rehospitalization after LVAD surgery.

Right Heart Failure

The HeartMate 3, like all LVADs, supports only the left side of the heart. When the pump dramatically increases the output of the left ventricle, the right ventricle suddenly faces a much higher volume of blood returning from the lungs and must keep pace. In some patients it cannot. Depending on how right heart failure is defined, estimates range from about 5% to 44% of LVAD patients developing this complication, and it is associated with worse outcomes.¹³ Predicting who will develop right heart failure before surgery has proven difficult. Existing prediction models have shown poor to modest accuracy when tested in different patient populations, partly because centers define right heart failure differently.¹⁴ Machine learning approaches using data from right heart catheterization and other preoperative measurements have shown promise in early studies, with one model achieving strong predictive accuracy, though broader validation is still needed.¹⁵

Rethinking Blood Thinners After Implantation

Every LVAD patient needs anticoagulation to prevent blood clots, and the standard regimen has long been a vitamin K antagonist like warfarin combined with aspirin. The ARIES-HM3 trial challenged the aspirin part of that equation. The trial found that dropping aspirin from the regimen in HeartMate 3 patients did not increase the risk of thrombotic events and led to a meaningful reduction in nonsurgical bleeding.¹⁶

A prespecified subgroup analysis of the same trial focused on patients who had traditional reasons to take aspirin, such as prior coronary stenting, bypass surgery, stroke, or peripheral vascular disease. Even in these patients, aspirin avoidance was safe and not associated with increased clotting, while it substantially reduced bleeding events.¹⁷ This is a shift in practice, because cardiologists have historically been reluctant to withhold aspirin from patients with atherosclerotic disease. The ARIES-HM3 findings suggest that the HeartMate 3’s low thrombogenicity changes the risk-benefit math enough to make aspirin unnecessary alongside warfarin.

The Surgery Itself

The standard approach for HeartMate 3 implantation is a median sternotomy, the same chest-opening incision used in many open-heart surgeries. However, for patients in whom a full sternotomy carries high risk, such as those who have had prior cardiac surgery with extensive scar tissue, implantation through a left thoracotomy is an alternative.¹⁸ A comparative study found that survival at six months and one year was not significantly different between the two approaches.¹⁹ The less invasive route is not yet the default, but it gives surgical teams flexibility for patients whose anatomy or surgical history demands it.

Daily Life and Quality of Life

Living with a HeartMate 3 involves carrying external equipment at all times. The controller is roughly the size of a small smartphone, and the lithium-ion batteries worn on a belt or in a harness provide hours of portable power. Patients can also plug into a home power unit overnight. Showers require waterproof dressing kits for the driveline exit site. Swimming and submersion in water are off-limits. Activities like driving may be possible after recovery, depending on the patient’s condition and local regulations, but contact sports and activities that risk impact to the driveline are avoided.

Despite these restrictions, functional capacity and quality of life improve substantially after implantation. In the MOMENTUM 3 trial, patients in both the HeartMate 3 and HeartMate II groups showed large improvements in the Kansas City Cardiomyopathy Questionnaire, a patient-reported measure of heart failure symptoms and physical limitations, within six months. Walking distance on the six-minute walk test also improved.²⁰ Longer-term registry data have shown that these improvements are sustained over time.²¹

The psychological side is harder to quantify. Follow-up research has found that both patients and their caregivers report moderate levels of anxiety and depression in the year after discharge. Sleep disturbances are common, partly from the constant hum of the pump and partly from caregiver vigilance about alarms. Couples frequently report that the device affects their intimate and sexual relationships. Caregivers in particular describe impairment in their social lives and difficulty maintaining their own self-care.²² These findings highlight that an LVAD decision involves the whole household, not just the patient.

Can the Heart Recover Enough to Remove the Pump?

By unloading the failing left ventricle and restoring normal blood flow to the body and to the heart muscle itself, an LVAD can trigger a process called reverse remodeling. The heart may shrink back toward normal size, and its contractile function may improve. Most LVAD patients show some degree of this structural improvement, but only a small fraction recover well enough to have the device removed. Among those who do, relapse back into heart failure is common, which has led researchers to speak of “remission” rather than “recovery.”²³

One prospective multicenter study specifically designed to maximize recovery used a protocol combining LVAD support with aggressive medical therapy in patients with nonischemic dilated cardiomyopathy. About 50% of patients who completed the protocol were explanted within 18 months, and after explantation, survival free of needing another LVAD or transplant was about 90% at one year and 77% at three years.²⁴ These numbers are encouraging but apply to a carefully selected population. For the broader LVAD population, recovery-based explantation remains the exception rather than the rule.

Use in Children and Adolescents

The HeartMate 3 was designed for adults, but it has seen growing off-label use in older children and adolescents whose body surface area is large enough to accommodate it. A single-center pediatric series reported outcomes in 15 patients with a mean age of about 11 years and body surface area ranging from 0.85 to 1.76 square meters. Most had dilated cardiomyopathy. Overall one-year survival was 87%, and 60% of patients were successfully transplanted during a median follow-up of 17 months. Complications included stroke in two patients and driveline infection in two others.²⁵ These early results suggest the device can work in larger pediatric patients, but the data are still limited to small case series.

Cost-Effectiveness and Access

LVAD therapy is expensive by any measure. The device, the surgery, the intensive care stay, and the years of outpatient follow-up add up to a substantial cost. A health-economic analysis from the United Kingdom estimated that HeartMate 3 destination therapy in transplant-ineligible patients cost about £47,000 per quality-adjusted life year gained compared with medical therapy alone, with a high probability of being cost-effective at the UK’s conventional threshold. For the sickest patients, those dependent on intravenous heart-failure drugs, the cost per quality-adjusted life year was somewhat lower. For less acutely ill ambulatory patients, it was higher.²⁶ These numbers are specific to the UK healthcare system, but the general pattern holds elsewhere: the sicker the patient population, the more favorable the cost-effectiveness looks, because the alternative, remaining on escalating medical therapy with very poor survival, is both costly and ineffective.

Eliminating the Driveline

Ask any LVAD patient or surgeon what they would most like to change about the device, and the driveline usually tops the list. The percutaneous cable is the leading source of infection and one of the biggest constraints on daily life. Several engineering groups have been developing wireless power transfer systems that would allow the pump to receive energy through the intact skin. One approach, coplanar energy transfer, has already been used in a first-in-human case. The system uses two large implanted rings in a coil-within-a-coil design to achieve robust wireless power delivery, paired with an implantable battery that provides over six hours of untethered operation.²⁷ If wireless power systems prove durable and reliable in larger studies, they could fundamentally alter what it means to live with a mechanical heart pump, removing the infection pathway and allowing activities like swimming that are currently impossible.