Preimplant Patient Status

In general, mechanical circulatory support is indicated when medical therapy has failed. The phrase “when medical therapy has failed” can be interpreted in different ways and is dependent on the experience of the program. Boyle and others used the Interagency Registry for Mechanically Assisted Circulatory Support (Intermacs) system to study adult patients with cardiogenic shock (stage 1) and progressive decline on inotropic support (stage 2). They found that these two groups had worse outcomes after VAD placement than those who were implanted earlier in the disease process (stable on inotropes) (Table 46.1) [11, 12]. It has also been shown that adults with end‐organ dysfunction, when undergoing device placement, have increased mortality [12–14]. In pediatrics, similar results were shown in the Berlin Heart EXCOR trial, as patients with preimplant renal dysfunction or increased bilirubin had a 4‐ to 7‐fold increase in mortality compared to those who did not [4]. Despite data supporting earlier implantation, implantation in cases of cardiogenic shock was common in pediatrics. Data from the Berlin Heart EXCOR trial database showed that 44% of patients were Intermacs level 1 and a slightly higher percentage (57%) were in the compassionate use cohort [4, 12]. In 2016 data from Pedimacs showed that only 25% of our VAD patients were in Intermacs 1 status at implantation [15].

While the optimal timing of implantation is subject to debate, it is clear that increasing center experience may fundamentally alter the clinical choices and outcomes as the mechanical support team decision‐making process matures. Currently, our institution begins evaluating patients in heart failure who require one inotrope and have some evidence of end‐organ dysfunction (i.e., intolerance of enteral feeds or the inability to wean from the ventilator). This decision must also assess the safety profile of the VAD being considered. For example, an infant weighing 5 kg who is extubated and gaining weight via enteral feeds on milrinone would be observed due to the potential risk of a Berlin Heart EXCOR in this age group, and because VAD therapy at this size does not allow the patient to be discharged. In contrast, an adolescent with dilated cardiomyopathy (DCM) dependent on milrinone who is eating would be considered a candidate for a VAD given the overall proven benefits demonstrated with CF‐VADs within this age group (Table 46.2). Thus, as centers gain experience with VAD implantation and more of our patients have the opportunity to be supported with advanced/durable devices allowing for discharge, earlier implantation will be the trend.

Intent of Device Therapy

Device therapy as a bridge to transplantation is the most common indication for pediatric mechanical support. Children bridged to recovery with advanced heart failure such as myocarditis, postcardiotomy myocardial dysfunction, or postarrest cardiac dysfunction also are commonly encountered. Although often difficult to determine, the possibility of recovery and the intent of the “bridge” must be factored into the choice of which device to use for support. Many institutions divide devices into short‐term and durable support options (Table 46.3). We define short‐term assistance, including ECMO, as support that will be used for approximately two weeks. This time frame is based on ECMO data that has shown an increase in the occurrence of complications after two weeks [14]. We no longer treat isolated heart failure with ECMO. Instead, we reserve its use for ECMO cardiopulmonary resuscitation (ECPR) or a patient with heart failure and pulmonary insufficiency who cannot be ventilated without damaging the lungs.

Short‐term assistance is better defined by the intent to treat and implantation strategy. Short‐term support is usually initiated using cardiopulmonary bypass (CPB) cannula without the use of cardiopulmonary bypass. Placement of long‐term cannulae using CPB and significant manipulation of the heart counteract the intent of short‐term support, which is a quick nontraumatic procedure to provide circulatory support; thus, to avoid bleeding and the use of blood products. Short‐term support is used because the etiology of the patient’s heart failure or neurologic status is unknown, and this strategy provides immediate support for the patient without committing to a long‐term device (Figure 46.1). Also, short‐term support is used if the etiology of the patient’s heart failure is inflammatory, such as myocarditis or acute allograft rejection, and we expect it to resolve in less than two weeks. If recovery does not appear to be an option, a more durable long‐term device will be chosen. Another reason to apply short‐term support is when the patient, who may be well known by the team, has a precipitous decline and is at Intermacs level 1. To optimize outcomes, clearly this type of patient and their end organs need to be resuscitated before considering long‐term VADs or other therapies. Bridging with short‐term support allows implantation of a durable VAD at an Intermacs level 2 or 3, which has been shown to have a survival benefit [15]. Additionally, the time afforded by short‐term support allows the team to initiate a full transplant evaluation, as transplant candidacy continues to play an integral role in device choice for pediatric patients and is mandatory for children less than 25 kg for whom device options are more limited. Over the long term, given the scarcity of organs, use of continuous‐flow devices and the development of miniaturized devices will be commonly employed as pediatric chronic therapy. In the current era, however, transplantation remains the most common goal for durable devices.

Underlying Diagnosis

The etiology of advanced heart failure in pediatrics is diverse and can include acute myocarditis, dilated cardiomyopathy, and various forms of CHD. It is important to consider the underlying disease etiology when determining patient selection, since this may help guide the type of assistance needed, as well as identify potential risks and the likelihood of a successful outcome.

Acute disease processes such as fulminant myocarditis or acute rejection in a heart transplant patient are the prototypical examples of rapidly progressive heart failure with circulatory shock that result in high mortality if not promptly recognized and treated. Historically, ECMO was the mainstay of mechanical support in these patients. However, short‐term VADs, placed percutaneously or surgically, have also been used successfully in this population and currently this is the trend in treating these patients [16–18]. Similarly, heart transplant recipients who develop severe graft dysfunction from acute rejection may benefit from temporary VAD support. Unlike myocarditis, the long‐term prognosis for transplant patients with severe dysfunction from acute rejection is guarded. In a study by Morales and colleagues, the survival to discharge rate for heart transplant recipients with acute rejection treated with mechanical circulatory support was only 71%, with one‐year survival of 50% and three‐year survival of 38% [19].

Patients with chronic diseases, such as CHD or cardiomyopathies, may benefit from more durable VAD support. Patients with DCM have been bridged successfully to transplant with both pulsatile and continuous‐flow devices, although differences in survival and morbidities exist between continuous and pulsatile devices [4, 15, 20]. Patients with hypertrophic or restrictive cardiomyopathies can be challenging to support with VADs given severe diastolic dysfunction and a relatively small ventricular cavity, which may impair VAD filling. However, unlike dilated cardiomyopathy, there is little medical management that can be used to treat patients with end‐stage severe restrictive physiology. Given these issues, standard criteria for device implantation for patients with DCM may be altered. Despite decreased support options for these patients, successful outcomes have been achieved with atrial cannulation and the total artificial heart, which is particularly effective for severe biventricular diastolic dysfunction [20].

Patient Size

Patient selection must consider the child’s size and the current VAD technology available that will fit in the child. Currently, the most frequently used durable devices are the Berlin Heart EXCOR, the HeartMate II (Thoratec, Pleasanton, CA, USA), the HeartWare HVAD (Medtronic, Framingham, MA, USA), and the total artificial heart (SynCardia Systems, Tucson, AZ, USA). Obviously, as VAD technology improves and devices become smaller and more reliable, the criteria for selection will evolve.

The only durable VAD option for infants weighing between 5 kg and 10 kg is the Berlin Heart EXCOR. Karimova et al. reported a 91% survival rate in a single‐center study of 11 children weighing <10 kg without CHD and supported with the Berlin Heart EXCOR [21]. In a study of 97 children weighting <10 kg, Conway and coworkers demonstrated a successful outcome (i.e., transplant, weaning off VAD with good neurologic outcome) in 57% of patients [22]. Importantly, this was significantly less than in children weighing >10 kg (57% vs. 83%, p<.001). The only independent risk factors for death on multivariable analysis were pre‐existing CHD and hyperbilirubinemia (>1.2 mg/dL). Approximately 70% of children who died had an elevated bilirubin before VAD implantation, further stressing the importance of early implantation before end‐organ dysfunction. The results of children weighing <5 kg were particularly grim, with only 27% having a successful outcome. It should be noted that many of the patients within this weight category were compassionate use patients and were more likely to be supported on ECMO and have worse end‐organ function. Also, many patients who had a failed palliation for CHD went on to ECMO, and had a VAD as salvage therapy. This cohort is known to have survival expectations of 10–20%. Although challenging, a child weighing <10 kg with DCM on milrinone, with minimal end‐organ dysfunction and who is intubated, should expect to be supported to transplant successfully without a stroke in an experienced program >70% of the time. Thus, reported survival should be interpreted with caution.

For children weighing more than 10 kg and up to 25 kg, the Berlin Heart EXCOR continues to be used successfully and is the only device to show consistently good outcomes for these patients. However, recently there has been an important paradigm shift toward CF‐VADs such as the HVAD Medtronic. Thus, for patients who are ≥25 kg, the overwhelming trend is to implant a CF‐VAD for durable support if the size of the child permits, due to their substantially lower complication profile, ability to go home, and the potential for chronic therapy. There is a great deal of discussion about whether these stated advantages are true when the HVAD is implanted in patients who are <25 kg and down to a body surface area (BSA) of 0.6 mm². A recent but small multicenter experience reported favorable outcomes in children with a BSA <1 m² (median 0.8 m², range 0.6–0.9 m²) and a weight of <23 kg (median 18.6 kg, range 13.5–23 kg) [23]. However, it is clear that there also have been many disappointing outcomes in patients weighing <25 kg with HVAD implants. One series showed a 100% stroke rate; thus, consistent efficacy and safety in this cohort weighing <25 kg have not been demonstrated. For larger children and adolescents, a recent comparison between children and young adult patients receiving a CF‐VAD demonstrated similar and excellent six‐month survival rates in both groups of 95% [24].

Patient Comorbidities

Advanced heart failure can be associated with multiple comorbidities, including renal dysfunction, hepatic dysfunction, pulmonary hypertension, and neuromuscular disease. Multiple scoring systems have been devised to help characterize these patient comorbidities. The most widely used scale for assessing patient acuity in adult VADs is the Intermacs scale [25]. When compared to adults, pediatric patients are more commonly at Intermacs 1 (27% vs. 14%) [4, 15]. Interestingly, the percentage of pediatric patients implanted with devices who are Intermacs 1 appears to be decreasing. Using the Berlin Heart EXCOR data published in 2013, 57% of patients were Intermacs 1 [4]. While in 2016, the percentage of Intermacs 1 patients had decreased to 26% per the Pedimacs dataset. This likely reflects the trend toward early implantation with VAD program maturation [15].

Approximately 25–55% of pediatric patients with VADs will have renal dysfunction at the time of implantation [26]. Interestingly, patients with baseline renal dysfunction have been shown to have worse post‐VAD outcomes (Figure 46.2). There was a >70% mortality within one month of VAD implant among patients with an estimated glomerular filtration rate (eGFR) <30% supported with a Berlin Heart EXCOR, in comparison to approximately 90% survival in patients with a normal eGFR [4]. While renal dysfunction predicts a worse outcome, the eGFR typically improves following VAD therapy, suggesting that much of the renal injury seen in these patients is reversible [27].

Hepatic dysfunction from advanced heart failure appears to be driven by venous congestion and impaired blood flow that typically can be reversed with hemodynamic improvement if fibrosis has not occurred [28–31]. Studies in both pediatric and adult patients have shown that hyperbilirubinemia and hypoalbuminemia are predictors of adverse outcomes [32–35]. As previously mentioned, hyperbilirubinemia with a bilirubin >1.2 mg/dL was associated with an approximately 65% mortality within one month for patients supported on the Berlin Heart EXCOR [4]. It is important to note that biomarkers may provide important information regarding hepatic function; however, liver fibrosis does not often correlate with these markers and thus should be evaluated specifically by other modalities. Similarly, liver dysfunction associated with Fontan physiology is a distinct entity and requires separate investigation.

Pulmonary hypertension and elevated pulmonary vascular resistance (PVR) are well‐known complications of advanced heart failure. VADs may help improve these abnormalities by unloading the left ventricle and decreasing left atrial pressure [36–38]. However, it remains unclear whether this benefit applies to children with CHD and, in particular, those with palliated single‐ventricle physiology [39, 40]. It is generally accepted that placement of a VAD in a patient with cardiomyopathy or corrected biventricular circulation and high PVR (even those who are minimally reactive) will have a significant reduction in PVR with good left ventricular assist device (LVAD) support [41]. The timing of this improvement is still not well characterized, but ranges from weeks to many months.

Neuromuscular disease was once considered a contraindication to transplant and mechanical support; however, with recent success in this population, these patients are being considered for chronic VAD therapy. In particular, adolescent boys with Duchenne muscular dystrophy (DMD) have shown improved outcomes given advances in chronic and respiratory care, such that heart failure is now the main driver of morbidity and mortality [42]. This has led some centers to implant durable CF‐VADs in patients with DMD and advanced heart failure [43, 44]. Since VAD use is still in its infancy within this population, patient selection remains at the discretion of the center. Programs that have supported these complex patients agree that VAD support prior to the onset of end‐stage disease when these young men can no longer use any of their extremities and/or are on constant positive‐pressure respiratory support appears to be a good management strategy for these patients.

Extracorporeal Membrane Oxygenation

ECMO is typically indicated in an acute setting, because the circuit can be set up rapidly, cannulation is expeditious and less invasive, and it can be performed without a sternotomy. Traditionally it has been employed for acute indications such as fulminant myocarditis [16], postcardiotomy failure to wean from cardiopulmonary bypass [45], acute decompensation in the intensive care unit (ICU) post cardiotomy [46], graft rejection following cardiac transplantation [47], or any sudden deterioration of hemodynamics where the ultimate outcome is still unknown. ECMO is often first‐line salvage therapy, which allows immediate end‐organ perfusion and tissue oxygenation. Whether ECMO will be a bridge to myocardial recovery depends on the underlying condition. If recovery is going to occur, it should be expected in a few days to two weeks [48]. When patients cannot be supported by mechanical ventilation alone, ECMO is indicated for cardiac arrest and respiratory failure. Traditionally ECMO was often used in acute isolated heart failure, but this practice is changing now that more specific forms of mechanical support (i.e., temporary centrifugal VADs) are being introduced.

Certain factors may suggest that ECMO will not be a bridge to recovery. These include failure of return of ventricular function within 72 hours of institution of support [49–52], serum pH, bicarbonate, and urine output by 24 hours of support; or a persisting plateau of troponin‐I levels beyond 48 hours after initiation of ECMO [50]. When any of these issues are present, other alternatives must be considered. Univariate analysis risk factors for early mortality on ECMO instituted after congenital heart repair include renal failure, extended periods of circulatory support (>72 hours), and prolonged cardiopulmonary resuscitation time prior to ECMO initiation (>45 minutes) [37]. Soon after ECMO initiation, left atrial pressure should be assessed, because it is usually high (unless there is known unrestricted left‐to‐right shunting) if left ventricular function is compromised. One must also quickly address high left atrial pressure if the lungs are expected to recover. This can be performed by a balloon atrioseptostomy (BAS) [53]; however, this will not decrease the left atrial pressure to zero. Therefore, if pulmonary edema is already present, it is recommended to do a direct left atrial cannula via sternotomy or thoracotomy.

Once functional myocardial recovery is not expected, ECMO may serve as a bridge to decision, a bridge to cardiopulmonary bypass to readdress residual congenital lesions after surgery, a bridge to VAD, and, in rare cases, a bridge to transplantation or retransplantation [54]. The latter should rarely be the intent, because survival is the ultimate goal, not transplant. It has been shown, unlike in the VAD population, that those supported on ECMO who reach transplant do significantly worse following transplant than those who were not on ECMO. The 2004 Extracorporeal Life Support Organization (ELSO) registry data showed that ECMO deployment for cardiac failure of any etiology across all age groups continues to be suboptimal, with survival to discharge rates of 38%, 43%, and 33% for neonates, children, and adults, respectively [55]. Because of these consistently poor results, the authors’ program as well as others will convert from ECMO to a temporary VAD once the lung function has clearly improved and the issue becomes isolated heart failure.

Short‐Term Ventricular Assist Devices

Short‐term VADs, most commonly CF‐VADs, have classically provided a mechanism to support patients with potentially reversible cardiac dysfunction such as myocarditis and postcardiotomy cardiac dysfunction (Table 46.3). For refractory cardiogenic shock, they may be used as a bridge to decision given the suboptimal outcomes associated with placement of durable devices in these patients [11, 14, 56]. As a referral center, one may also receive transfers of very ill patients who are not well known to the institution, but who clearly need support. Implantation of a short‐term device can be done quickly off cardiopulmonary bypass, allowing stabilization and time for the team to assess a patient’s etiology of heart failure and their candidacy for transplantation or chronic VAD therapy. There is limited, published pediatric experience with short‐term devices, though anecdotal experience suggest the use is growing [49, 50]. The para‐ or extracorporeal pump does not determine short‐term support, but rather the cannulas being placed. Long‐term cannulae placement (i.e., EXCOR cannulae) is a more involved surgery and may negate the advantages of temporary support (i.e., the rapid employment, minimal heart manipulation, avoidance of cardiopulmonary bypass, and blood products). When used as a temporary VAD, closure of the sternum rarely provides benefit for the patient. If the patient is awake and extubated, they should be either weaned from the VAD or bridged to durable support with more secure cannula (i.e., EXCOR cannula and continuous‐flow pump). The pump does not determine support intent (durable vs. short term); however, the approach to support does. The most commonly used short‐term VADs are CF‐VADs, which are able to support patients across the age/size spectrum (Table 46.3

Stay updated, free articles. Join our Telegram channel

Join