Previous reports have demonstrated that pump exchange from HMII to HM3 was feasible and safe; however, most clinical experiences were in Europe and the USA. There is a need for more information concerning Asian patients. To the best of our knowledge, this is the first case report presenting an Asian patient undergoing HMII pump exchange to HM3 successfully.

According to the literature review, nearly half of the patients received redosternotomy for HMII exchange to HM3. In our patient, we performed limited left anterior thoracotomy and lower partial sternotomy to avoid redosternotomy. Although the incision of pump exchange is mainly determined by the position of the pump and inflow graft, our procedure provides an alternative and limited approach to pump exchange.

Among the patients in the literature review, only our patient underwent HMII exchange to HM3 due to driveline damage. In our patient, the driveline damage possibly resulted from fatigue failure due to repetitive flexing and abrasion against the braided metal shield. We did not consider exchanging the HMII pump to HMII because of two reasons. The driveline design of the HM3 is better than the HMII, with a lower risk of driveline damage [15]. Additionally, it is hard for the patient to accept pump exchange to the same LVAD model, which did not work after only three years of use, instead of exchange to a new LVAD model.

This literature review showed that pump thrombosis was the most frequent indication of pump exchange, but only one (3.0%) patient encountered recurrent pump thrombosis after exchange. As the MOMENTUM 3 trial results showed that the HM3 group had a significantly lower risk of pump thrombosis than HMII [2], our review implied that the superiority of HM3 to HMII might not only be in primary implantation but also device exchange. However, this review also found four (12.1%) patients experiencing death or another pump replacement, comparable with the results of INTERMACS reports [1]. Therefore, much research is needed before a definitive conclusion can be drawn.

Infection is a common adverse event of LVAD, regardless of primary implantation or device exchange. The present review noted that infection was the second most common reason for pump exchange, consistent with the previous reports [1, 4]. Notably, four (12.1%) patients encountered pump pocket infections after surgery. HMII and HM3 have different characters in pump size and shape, location to place, outflow length, and insertion angle. With surrounding adhesive tissues, the old HMII pump pocket might become a dead space and a nidus of infection after pump exchange. Similar reasons could explain another case in the literature review, who received reoperation for inflow malposition on post-exchange day 7. Therefore, more attention should be paid to these potential complications after surgery for patients undergoing pump exchange to a different device.

In this case report, the patient successfully underwent pump exchange from HMII to HM3 and did not experience any hemodynamic adverse events or device malfunction during postoperative follow-ups of 12 months. The case demonstrated that it is safe and feasible to perform HMII LVAD exchange to HM3 through a limited thoracotomy and lower partial sternotomy for Asian patients.