Heart failure remains one of the leading causes of morbidity and mortality worldwide, driven by population ageing and improved survival after cardiovascular disease. Patients with advanced heart failure may require mechanical circulatory support, such as durable left ventricular assist devices (LVADs), including as destination therapy when heart transplantation is not feasible. Despite major advances in LVAD technology and the associated improvement in outcomes, driveline infection remains one of the main complications of this therapy, with a substantial negative impact on both quality of life and prognosis.

Over recent years, several device platforms have been developed, incorporating progressive technical refinements in their components. Previous studies have linked driveline infection risk to the design and physical properties of the driveline system, including trauma at the exit site caused by external or internal forces, as well as any factor that may predispose the exit site to microbial invasion. The CH-VAD (BrioHealth Technologies, Suzhou, China) is a fully magnetically levitated centrifugal LVAD that, compared with currently available devices such as the HeartMate 3, has a thinner driveline (3.3 mm vs 6.6 mm) and greater flexibility. This design is intended to reduce local trauma at the exit site and potentially lower the risk of infection. The device was first implanted in China in June 2017 and received approval in that country in 2021.

This was a multicentre, retrospective, observational study designed to assess the incidence of driveline infection in patients supported with the CH-VAD and to explore factors that may contribute to this complication. All patients who underwent implantation of this device between June 2017 and October 2024 at 9 centres in China were analysed. Data were collected on clinical characteristics, surgical procedures, and prevention and treatment strategies for driveline infection, and risk factors were evaluated.

A total of 181 consecutive patients were included, with a median age of 56 years, and 86% were men. During a median support duration of 1.6 years, driveline infection occurred in 20 of 181 patients (11.0%). Twenty-six events were documented, corresponding to a rate of 0.07 events per patient-year, among the lowest reported for durable LVAD therapy. The cumulative incidence of first infection was 2.8% at 6 months, 5.6% at 12 months, and 8.5% at 24 months. Of note, 20 of the 26 events yielded positive cultures, with Staphylococcus aureus as the predominant pathogen (65%). Most infections were superficial (65%), and treatment followed a stepwise approach: local care alone, antibiotics for 2-4 weeks when there was no deep infection, and 2-3 months when deep infection was present; some patients also required debridement, and 1 case underwent driveline relocation. Most events resolved successfully (88.5%), and deaths related to driveline infection were infrequent (3.8%), including 1 case complicated by mediastinitis.

Two findings are particularly relevant. Frequent dressing changes (>5 per week) were significantly associated with a higher risk of driveline infection (p = .022). In addition, complete subcutaneous burial of the velour portion of the driveline was associated with a lower infection rate (7.4% vs 26.3%). Moreover, implants performed before 2022 showed a higher infection rate (27.3% vs 7.4%), probably reflecting the learning curve, increasing standardization of care, and longer follow-up.

The authors conclude that patients treated with the CH-VAD experienced an encouragingly low rate of driveline infection, regarded as one of the lowest reported for durable ventricular assist devices. These findings have implications not only for clinical management, but especially for future driveline design in ventricular assist technology.

COMMENTARY:

Driveline infection remains the Achilles’ heel of durable ventricular assist devices. Despite the remarkable improvement in haemocompatibility and survival achieved with centrifugal fully magnetically levitated pumps, the driveline continues to represent an access point for infection and, therefore, a source of readmissions, prolonged wound care, and occasionally sepsis or even mediastinitis. Against that background, this study from 9 centres in China, including 181 patients treated between 2017 and 2024, delivers a message with very practical implications: prevention of driveline infection is not just a matter of protocol, but also of cable mechanics and design.

This work highlights 3 important points: 1) very low driveline infection rates compared with those reported for durable LVADs; 2) an ultrathin, highly flexible driveline design supported by previous literature linking larger diameter and greater stiffness with increased infection risk, as well as by experimental studies on driveline mechanical properties; and 3) the identification of 2 aspects of driveline care associated with better outcomes, namely less frequent dressing changes and complete burial of the velour segment beneath the skin.

It may be too ambitious to attribute the low infection rate solely to driveline design rather than to the combination of design and patient management protocol. Even so, these findings provide several useful lessons for daily clinical practice. First, the driveline is not merely an accessory, but an integral component of the device with clear clinical and prognostic implications, and it should not be overlooked in future technological advances. Second, less may sometimes be more in driveline wound care, underscoring the importance of meticulous management while avoiding unnecessary manipulation that may facilitate microbial entry. Third, available data appear to support complete subcutaneous coverage of the rough or velour portion of the driveline, while also emphasizing the need to use appropriate antiseptics and materials so as not to damage this structure.

Among the strengths of this study, I would emphasize the reasonably large sample size for a relatively new device, its multicentre design, and the longest reported follow-up reaching 7 years. Its careful statistical approach and the reference to the DESTINE classification, together with comparison against HeartMate 3 experience, further strengthen the quality of the results. At the same time, its limitations should not be overlooked, particularly its retrospective nature, the absence of a control group, and the limited statistical power imposed by the small number of recorded events, with 26 events in 20 patients. What seems most relevant to me is the potential confounding surrounding the comparison before and after 2022, where the learning curve is intertwined with changes in implantation technique, wound-care protocols, and patient follow-up. The authors acknowledge this point, and therefore the lower rate of driveline infection in the later period cannot be interpreted as a spontaneous improvement without important caveats.

In conclusion, the low driveline infection burden observed with this newer, thinner, and more flexible design is highly interesting, but it should still be regarded as an encouraging and promising signal rather than definitive proof of superiority over currently established driveline designs.

REFERENCE:

Wang X, Zhou X, Chen H, et al. Low Driveline Infection Rates in Patients With a Novel Fully Magnetically Levitated Ventricular Assist Device. Eur J Cardiothorac Surg. 2025;67(12):ezaf365. doi:10.1093/ejcts/ezaf365