**5. Special anatomic congenital considerations**

Another challenge in utilizing MCS devices in children or adults with congenital heart disease is the incredible amount of variation that occurs either at birth or after multiple palliative operations. We discuss some of these specific examples to highlight the challenges and the limited data available on current usage. Although this appears to be increasing, the practical application of MCS to these challenging patient cases will require more experience and long-term follow-up.

#### **5.1 Single ventricle patients**

Single ventricle patients who undergo staged palliation requiring partial or complete cavopulmonary anastomosis present a unique challenge to implementation of MCS. These challenges include technical/anatomic differences in these patients, as well as physiologic. However, this population is also at significant risk for heart failure during the interstages between Stage I (Norwood or Hybrid), Stage II (bidirectional Glenn), and Stage III (Fontan operation). For example, the authors from Kansas City describe a case report supplying MCS in a superior cavopulmonary patient as a bridge to transplant in an infant [23]. In **Figure 1** [23], the authors illustrate cannulation in the reconstructed aortic root (Damus-Kaye-Stansel) and systemic ventricle for a failing Glenn patient. They describe the patient's changes in physiology while maintained on support and prior to successful transplantation after one year. They conclude that successful long-term support is feasible and ECMO in parallel can be instituted as a strategy for short-term pulmonary support [23].

#### **Figure 1.** *Ref. [23]. Reprinted with permission from Elsevier under CC-BY-ND-ND license.*

The Fontan operation is typically the final staged palliation for single ventricle patients that connects the systemic venous return to the pulmonary circulation (total cavopulmonary anastomosis). In Ref. [24], a broad description of the background, physiology, and details of different types of Fontan surgically created circulation can be reviewed for building a knowledge foundation. The failing Fontan is another category of research that elucidates the inherent challenge in these patients of ultimately needing a heart transplant [25]. This patient population can be particularly medically challenging due to manifestations of chronic Fontan circulation including liver dysfunction, valve regurgitation, myocardial dysfunction, collaterals, plastic bronchitis, and protein-losing enteropathy. With the addition of anatomic and physiologic challenges, inserting MCS devices into this population is currently under review. Fontan failure can be multifactorial with cardiovascular factors including ventricular dysfunction [6]. Once failure symptomatology starts, transplant waitlist mortality is high [6]. Several limited options exist for supporting failing Fontan physiology including VAD implantation [6, 26].

Using VADs addresses myocardial failure, but will not reverse cavopulmonary failure. Currently, there are no pumps specifically designed for Fontan cavopulmonary circulation. Implantation of these devices may be anatomically challenging due to prevalent under development of one ventricle – leading patients down a singleventricle pathway. Despite these challenges, some bench research has supported

using of the VAD successfully in failing Fontan circulation. Also, the HeartMate 2 supported a sheep for 2 hours when placed in the cavopulmonary graft. Implantation of the HeartWare HVAD in three Fontan patients for 5 to 9 months were successful bridge to transplant [26]. However, in 2021, the HeartWare device was recalled and not currently available for implantation [5]. One case demonstrated the Berlin Heart provided biventricular support for a failing Fontan circulation [26]. HeartMate 3 implantation in a CHD Fontan patient is usually performed after multiple previous staged palliative operations [6].

One of the physiologic challenges after insertion of VAD into the ventricle is off-loading, but can also concomitantly increase Fontan pressures, which can lead to a worsening clinical situation. It does appear that continuous devices are more beneficial than pulsatile devices given the reliance of the circulation on passive pulmonary flow [26]. Careful consideration is needed of the underlying etiology for a failing Fontan, such as ventricular failure, elevated pulmonary artery pressures, or enddiastolic pressures.

In Ref. [6], the Society of Thoracic Surgeons created a group dedicated to observing the outcomes of Fontan patients supported with VADs from 2012 to 2019 using large mechanical assist device registries (Pedimacs and Intermacs). Retrospectively, they found 55 Fontan patients with a median age of 10 years and 27 kg who underwent VAD implantation. Highlighting the trend of more device implantation and growing experience, the equivalent number of VADs were implanted in the last year of observation compared to the previous 5 years (28 vs. 27; p = 0.01, from 2018 to 2019 and 2012–2017, respectively). In addition, the later era had a higher pre-VAD glomerular filtration rate, perhaps demonstrating earlier implantation. Overwhelmingly (89%) of devices were placed to support the systemic ventricle as left-sided VADs. The overall positive outcome (alive, transplant, or recovery) was demonstrated in 81% at 6 months with median length of support at 3.8 months. Fifty-eight percent of all mortality occurred during the first month. Demonstrating the length of duration VAD support can offer, five patients received support for greater than 1 year. Adverse events such as stroke, pump thrombosis, and bleeding ranged from 4 to 9% (1.4 to 3.3 out of 100 patient-months). Important implantation considerations in Fontan patients includes perioperative care, aortopulmonary collaterals, pulmonary vascular disease, or end organ dysfunction can coexist [6].

#### **5.2 Surgical modifications**

Currently, some alternative surgical techniques to implanting devices through a median sternotomy and placing the device intrapericardially are under investigation. A minimally invasive approach using two anterior thoracotomies or hemisternotomy are being explored with multiple potential advantages. Because it may be difficult to align the inflow cannula vertically, so that it lies parallel to the interventricular septum, another implantation technique is to divide the left hemidiaphragm to create a pump pocket, as described by reference [9]. Surgical modifications of tunneling the driveline in under-developed abdominal walls of smaller children may decrease an already elevated risk of driveline infections in this population by off-setting the insertion through the rectus abdominis muscle [9]. Resecting papillary muscle that may cause mechanical obstruction, reorienting the direction of the outflow, and padding the pump housing against the chest to try and decrease pain with a buffer are other surgical techniques described when treating smaller congenital heart disease patients [4].
