One of the parameters most commonly used to guide the selection of the right treatment for every patient is the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) classification (◘ Table 4.1), a risk stratification tool for durable MCS support candidates aiming to normalize the outcomes of the implanting centers in the US to the preoperative patients’ clinical conditions.

The classification appears extremely easy to apply and aims to overcome the wide spectrum of the possible preoperative clinical scenarios. Meanwhile the INTERMACS classification was adopted by the whole heart failure community and is now accepted worldwide.

The classification appears extremely easy to apply and aims to overcome the great spectrum of possible preoperative clinical scenarios. Meanwhile the INTERMACS classification was adopted by the whole heart failure community and is now accepted worldwide. The classification has been shown to be effective to evaluate the short-term mortality risk of a VAD patient as well as to identify adverse events associated with different INTERMACS profiles. As a consequence of the fact that patients in INTERMACS profile 1 (“critical cardiogenic shock”) experience higher mortality (from 65% up to 76% at 1 year) and morbidity, a shift was noticed from implanting durable VADs in patients with INTERMACS profile 1 [2] towards patients characterized by a lower acuity and severity of illness (30% implants in INTERMACS class I in 2008 to only 15% in 2013) [12]. Only durable VADs and TAH go into the INTERMACS registry. Unfortunately, this leaves behind the possibility of outcome analysis of TCS, mainly used for bridge-to -decision applications. In order to lower the costs and avoid futile implantations of expensive durable VADs the strategy of bridge-to-decision with TCS is becoming widely adopted within our community [10, 13].

To even better stratify the VAD population and evaluate concomitant risk factors, three modifier factors (TCS, arrhythmia, and frequent flyer (FF)) were successively added [14]. The TCS modifier is meant to be assigned to patients in profiles 1–3 who are supported with either with an intra-aortic balloon pump (IABP), an extracorporeal membrane oxygenator (ECMO), or a temporary ventricular (right or left) support (TCS) prior to their durable VAD implantation. The arrhythmia modifier can be applied to any profile and classifies patients with recurrent ventricular tachyarrhythmia that evoke clinical compromise. Finally, the FF modifier applies to patients in profile 3 at home on inotropes and patients in profiles 4–7. The FF modifier denotes a patient with frequent emergency department visits or hospitalizations for heart failure symptoms (◘ Table 4.1).

When IABP is removed from the TCS definition, the TCS modifier conferred >40% higher mortality [14]. The strategy of using TCS, might save costs in patients with otherwise durable VAD implantations, but can end up in a very complex management with multiple procedures and a predicted long ICU stay. Looking at the accuracy of prediction, the TCS modifier should probably only be applied only in patients supported with ECMO or temporary VAD support and should be always assigned to a profile 1 TCS. The real discriminant to clarify the effectiveness of such a strategy is to determine the percentage of patients receiving a TCS and being actively withdrawn from support, in order to establish the real amount of resources spared. Finally, addressing INTERMACS I those patients with a bridge-to-decision approach has the potential to reduce the higher mortality rate of this challenging patient population.

Arrhythmias as cause of decompensation should be always adequately addressed and eventually treated before or during LVAD implantation if its resolution is not deemed as a goal trough wall-stress reduction due to LVAD implantation. The clinical profile of patients with the arrhythmias was that of sicker patients, requiring more vasopressors and ventilator and dialysis support than patients without high-burden arrhythmias. However, arrhythmias may limit the impact of preoperative interventions such as IABP and inotrope support thus warping the accuracy of INTERMACS levels.

The FF modifier portended worse outcome beyond that of INTERMACS profile and other known risk factors; unfortunately, only 47% of patients with profiles 3–7 had the FF designation recorded, and this reduces the quality of prediction of such a modifier. However, these data should mandate an indication to MCS implantation when a patient does not recover an acceptable clinical stability after a first treatment with inotropes. A second episode of decompensation during the next 6 months should be a red flag indicating prompt MCS implantation.

Today, INTERMACS data fully support MCS implantation with durable devices when the patients are in classes I up to III. However, the feared increased risk of mortality and of adverse events in the lower INTERMACS classes dampened the use of implantable LVAD in such patients so that, despite the maximum clinical benefit, LVADs have been less commonly implanted in these settings.

It should be noted that the clinical results and the consequential benefit in this setting were realized in an era in which the largest amount of pumps implanted was pulsatile [2] and indeed not necessarily these data could be replied in the era of continuous-flow pumps. In the 2013 report the incidence of right ventricle failure (RVF) in INTERMACS I patients is significantly higher reaching an incidence up to 40% [2], thus mandating a careful selection of LVAD candidate versus BiVAD and TAH [15]. Weighting the clinical benefits and the costs, the adoption of a short-term easy-to-implant device in class I has become a common pathway of treatment for such complex patients especially when a complete neurological evaluation is not available and the metabolic status impedes to establish the reversibility of clinical conditions [16–19].

.

On the other hand, the excellent outcome reached by LVAD implantation in the last registered trials and in the real-world practice might be comparable to midterm results of heart transplantation in the era of marginal donors [20], and this is stimulating to plan studies in order to expand LVAD implantation to the higher INTERMACS classes [8, 21]. Until now, the high cumulative risks of complications causing re-hospitalization during the first year [2] and the amelioration of the outcomes of medical treatment (ROADMAP) prevented the expansion to less sick patients. Long-term data from a trial on medical management of patients suffering of end-stage heart failure (MEDAMACS) are awaited to further define the benefit of MCS in such patients and to identify clusters of patients with more aggressive clinical course beside a high INTERMACS profile.

Preliminary data recently available from the MEDAMACS Registry of patients in INTERMACS 4–7 showed a high overall event rate in the entire study [21, 22]. Of the total study population of 144 patients, 75.7% were alive without a transplant or LVAD, 11.8% had died, 4.9% received a transplant, and 6.3% had an LVAD implantation. The high mortality rates were driven by patients that were transplant/DT-LVAD ineligible. Transplant/DT-LVAD ineligible patients had substantially higher mortality rates (23.3% versus 8.0% in DT-LVAD eligible and 5.9% in transplant eligible, p = 0.02), a larger number of patients were requiring inotropes, and greater number of re-hospitalizations. These data clearly suggest that there is still an important unmet clinical need (>30% of patients in INTERMACS IV–VII) among those patients with a weak evidence for a VAD implant according to the guidelines, but with an overall bad final outcome. In the registry, clinically perceived frailty, social factors, and risk of nonadherence to therapies were the principal explanation of being labeled as transplant/DT-LVAD ineligible. These patients will be probably the new frontier for MCS in the near future waiting for a less invasive and easy to manage strategy of circulatory support such as cheaper and easy to handle pumps for partial support.

The management of End-Stage-CHF with MCS carries high costs due both to device direct costs and to indirect costs mainly related to the prolonged ICU/Hospital stay. Cost-effectiveness of MCS experienced a meaningful improvement in the recent era, partially due to the reduction of implant costs and also due to the improved knowledge of care with consecutive reduction in length of stay. The hospital costs decreased by 50% with the evolution from the pulsatile devices to continuous-flow device implantation [11, 23]. On the other end, the incidence of the pathology and also the socioeconomic weight in terms of re-hospitalization, with the relative consumption of human and economic resources, still represent a serious factor limiting the expansion of this technology.

LVAD therapy represents a mainstream therapy for advanced heart failure with a relative 75% reduction in cost/QALY over the past decade. The overall high costs for MCS is outweighed by the definitive clinical benefit of CHF patients in comparison to patients with other life-threatening diseases. The anticipated reduction of adverse events should further reduce the costs of MCS [24].

The British BTT experience was the initial cost-effectiveness analysis of continuous-flow LVADs, showing a cost per QALY of £258,913 ($414,275) (depending on the choosen device) [25]. This estimate outweighed of £30,000 for cost-effectiveness of the treatment, and the two determinants of poor cost-effectiveness were improved survival of advanced heart failure patients awaiting transplant and the high cost of the LVAD. Cost-effectiveness of LVADs from BTT trials is further complicated by the additional costs due to the subsequent transplantation. The more the follow-up is long the more the difference in costs increase due to the different survival between the two cohorts.

Real-life experience showed comparable length of hospital stay, number of readmissions, and 1-year survival between patients receiving a continuous-flow LVAD as a bridge and those going straight to heart transplant.

A different scenario has to be considered for the implantation of TCS systems in the highest-risk patients in which a higher hazard of failure, the increased risk of complications, the increased costs due to prolonged ICU stay, and the hazards due to multiple procedures may lead the patients treated with a TCS both to unacceptably high costs and to a bad outcome. In this view TCS should be used as bridge to decision only for brief periods and principally as a strategy of optimization of the main treatment, upgrading to a durable device as soon as the patient reaches criteria to be implanted. Data from the Medicare population ineligible for transplant disclosed an average cost of treating advanced heart failure with optimal medical management (OMM) of approximately $180,000, spent previously during the last 6 months of life [26].

Cost-effectiveness analyses, when applied to orphan diseases or other end-of-life treatments, can be challenging, as the evaluation does not consider the innovative nature of medicine or the availability of an alternative treatment.

Finally, current data suggest that the cost-effectiveness of VAD therapy as DT is improving but has yet to achieve the goal of <$100,000 USD/QALY through a careful patient selection and management [23]. Moreover, the sicker the patient the higher are the adjunctive costs strictly dependent from the increased survival. Thus, if futile implantations encompass high costs that could be completely spared without treatment, effective implantations spare life that cannot be saved differently.

However, looking at the possible future benefits of MCS in congestive heart failure given the social impact of the increasing prevalence of this disease and the increasing length of life, when coupled with more robust data on long-term survival and quality of life, it is reasonable to anticipate that mechanically assisted circulation will ultimately achieve cost-effectiveness.

Nowadays the distinction between bridge to transplant policy and destination therapy may be justified principally by the strict historical relation between MCS and transplantation at the dawning of MCS. In the actual situation of organ shortage (worsened by the increased survival of patients treated with medical therapy and with MCS), MCS is increasingly used as a treatment for HF because of its prompt availability, leaving heart transplantation (HTx) as a solution for patients not amenable of MCS implantation and for patients with a failure of MCS. However, ideally, HTx should be reserved to the few recipients that could experience a follow-up long up to 20 years or a significant amelioration of quality of life and consider MCS for all the others. Moreover, the MCS has also changed the clinical course of patients affected by ES-CHF so that patients with a low likelihood of HTx before MCS may become ideal candidates after an effective mechanical support. In a paper of Uriel [27], it was well pointed out how the definition of the strategy (◘ Table 4.2) may be subject to repeated changes during the support of the patient, and this led in Europe to label a new therapeutic strategy, known as “bridge to candidacy” (BTC). Finally, today it could be more appropriate [28] to distinguish the “indication” to HTx from the “candidacy” to HTx, considering for the latter a timely MCS implantation in case a suitable organ is not available in a reasonable time, in order to prevent a further decline of patient’s conditions and miss even the possibility for a safe LVAD implantation.

Destination therapy (DT) is the strategy for patient not eligible for HTx, either due to age or comorbidities. Destination therapy is meant to be a permanent, lifelong form of left ventricular support, and it represents a growing indication, offering the greatest potential for improvements in HF morbidity and mortality. An intriguing perspective is to provide early unloading to prevent remodeling in a cardiomyopathy with hopes for recovery (i.e., myocarditis) [29].

Implanting a patient as bridge to transplant (BTT), the effectiveness in terms of end-organ recovery should be kept in mind as first issue, trying to choose a device capable to offer full support to the patient and to perform a safe HT. An extensive work-up has to be performed [10, 30] to verify if end-organ dysfunction may be reversible keeping in mind the difference between acute (shortly reversible), chronic (often reversible over a mid-time), and acute on chronic illness (often requiring a very long period to recovery).

An intriguing perspective is the “bridge to recovery” (BTR) strategy, that aims to provide an early unloading to prevent ventricular remodleing in a cardiomiopathy considered to be reversible (i.e., myocarditis) [30]. Destination therapy (DT) is the strategy for patient not eligible for HTx, either due to age or comorbidities. Destination therapy is meant to be a permanent, lifelong form of left ventricular support, and it represents a growing indication, offering the greatest potential for improvements in HF morbidity and mortality. When implanting a patient as destination therapy, the principal target is to choose the device capable to improve clinical conditions and possibly quality of life of the patients while reducing the incidence of complications [31, 32]. The ENDURANCE trial (still unpublished data) disclosed different kinds of complications between the two most commonly implanted pumps (Heart Mate II and HeartWare) that should be balanced with preoperative features of the patients (i.e., risk of strokes after HW in patients with history of hypertension). In the effort to tailor the approach to the patient characteristics, some centers prefer to implant a device with partial [33–36] support (i.e., CircuLite Synergy) or less complete but without the abdominal driveline (i.e., Jarvik) wishing to give a better clinical perception to the patient and trying to eliminate the risk of driveline infection during the long term. In this perspective the knowledge of temporal course of complications and the capability to anticipate the kind of complications with respect to the baseline clinical characteristics may become a new horizon to offer the best results to a variegate population of patients. Indeed, the choice of a device with higher shear stresses may increase the probability of bleeding complications over the long term especially in an elderly patient, and, on the other end, early implantation with pumps capable to give a lower output may be preferred to preserve patient’s right ventricular function and to keep arterial pressure pulsatility in order to protect from gastrointestinal bleedings [37, 38]. Development of aortic valve regurgitation should be avoided [39, 40]. In such elderly patients with mildly incompetent fibrotic aortic valve (or dilated aortic root), an additional care may be to choose a device capable to preserve flow pulsatility by specific softwares generating periodic flow changes [41, 42]. Recently, some interesting data regarding pulsatility emerged also from the analysis of the Evaheart cohort [43]. It has been postulated that the peculiar flow-pressure curve of the pump and its design is capable to best preserve pulsatility and its effect both on coagulative disorders (less gastrointestinal bleedings) [44, 45] and on right ventricular dysfunction [46]. Another factor to be evaluated is the width of the driveline: devices with a thinner driveline seem to account to less driveline infections [47].

In conclusion, when implanting an elective patient with the aim of destination therapy, it should be fairly discussed with the patient the choice and the risk both of the available models and of the surgical approach trying to give to the patient the best outcome with the lowest incidence of complications.

The underlying disease is an important factor when planning the implantation of a system of TCS or MCS. In case of TCS the principal aims are the lower biological impact to reduce complications and the left ventricle unloading to increase the likelihood of recovery. In fact, the most common indication of TCS today is bridge to recovery. An alternative strategy is represented by the bridge-to-decision policy: this solution is increasingly used as soon as the patient recovers an optimal metabolic and end-organ function and neurologic dysfunction has been excluded. The bridge to transplant with ECMO should be carefully considered looking at the data coming from ISHLT database as the short- and midterm survival is heavily impaired. The length and the quality of TCS may represent the real discriminant factor with respect to transplant results. ECMO is the most commonly used system for TCS as documented by a recent study from Bartlett [48], and the maximum chance of recovery is reached after minimum 6 days of support [49]. A study on ECMO [18] showed that an ECMO support longer than 14 days is associated with a bad outcome with respect to a quick transition to durable MCS. Efforts should be always spent to plan the next step when the patients is bridged with a TCS and an organ is not timely available. According to the data from the ISHLT, patients coming out from ECMO have a lower survival after HTx that is similar to the survival of TAH/BiVAD [2]. Adjunct of a left ventricular drainage seems to increase the likelihood of a recovery [50].

In the case of durable MCS, the underlying disease in the light of a possible recovery [20] is becoming more and more debated looking at the possibility to heal the heart and not to replace it. Intriguingly, in the era of cf-LVAD, the incidence of recovery appears reduced. It is not clear if it depends from a lack of arterial wave pulsatility or if it depends from a different clinical profile of the candidate for LVAD implantation today compared to the era of pf-LVAD. Certainly time of implantation is important if aiming to a possible recovery both in myocarditis and in idiopathic cardiomyopathy (the most commonly recovering underlying diseases). In this setting, also if some evidence of reversal of fibrosis has been reported [51], the likelihood of a recovery appears really unpredictable when a significant grade of myocardial fibrosis has been already established.

Care must be taken when choosing patients with restrictive and hypertrophic cardiomyopathy both for surgical issues and for an increased risk of complications due to the placement of the inflow cannula [52]. However, these recipients if carefully selected may really benefit from LVAD implantation, in consideration of the increased risk of primary graft failure associated with heart transplantation [53].

Another fascinating and growing indication to LVAD implantation that should be accurately assessed in the near future is the support of congenital adult cardiomyopathies. Patients with a univentricular physiopathology have been treated with an LVAD implantation from many surgeons after the first report in 2005 from Frazier [54]. Numerous small series are showing the possible role of LVAD implantation and also of TAH in such a setting. The burden of the treatment of a failing Fontan is another big issue to approach. A number of patients with surgically corrected or palliated congenital heart disease are left with abnormal anatomy that makes VAD implantation difficult, impossible, or even unpredictable. Artificial conduits, valvular insufficiency and residual shunts may portend complications within the inflow or outflow of the device. Repair residual defects and then place a VAD or a biventricular assist device (BiVAD) is an attracting solution [55], but the morbidity/mortality profile will be then certainly higher compared to an isolated VAD placement, thus leading many authors to prefer directly a TAH solution [56].

LVAD implantation is based on the assumption of a normal right ventricular function, capable to maintain a sufficient preload warranting an optimal filling of the left ventricle. The capability to predict postoperative right ventricular failure (RVF) is one of the trickiest evaluations in the field of MCS: careful evaluation of preoperative parameters, timely diagnosis, and perioperative management are the key. Unplanned postoperative temporary RVAD/ECMO implantation for right ventricular failure may be a catastrophic complication with a significant impact on the length of postoperative ICU stay and mortality [57].

Without entering into specific considerations that will be explained further in this book, there are several factors that have to be considered when evaluating right ventricular function in a possible candidate to LVAD implantation:

Many scores try to predict adequately the incidence of RVF, but scores alone shouldn’t contraindicate implantation of an LVAD, being more a tool to address a more careful management of right ventricle from preoperative to postoperative course.

In the era of CF-LVAD, the incidence of delayed RVF is increased [58–60], partly due to different clinical profiles of the patients implanted nowadays, partly due to the absence of pulsatile flow, with a consequent dampened pulsatility in the right sections [61]. The adoption of solutions to preserve aortic valve opening and systemic pressure pulsatility try to address such an issue [38, 39].

In case of preoperative right ventricular dysfunction, right ventricle function has to be carefully evaluated in the attempt to distinguish reversibility of right ventricular dysfunction. The different possible solutions in case of concomitant right ventricular failure are shown in ◘ Table 4.3.

The fear of the absence of a backup circulation and the dimension of the device (lately resolved by the 50 cc TAH fitting approximately to all adults and small adults) is one of the main factors limiting expansion of TAH technology nowadays. This solution however still carries a great simplicity of management and the highest number of patients gaining HTx (looking at the seventh ISHLT report 56% versus less than 20%), thus appearing an effective solution in the setting of BTT. The safety of the transplant procedure when respecting the rule for a safe sternal closure of such patients is another additional argument in favor of TAH implantation [62]. The noise and the quality of life with this approach are one of the major concerns regarding the future development of this technology.

The choice between these possible approaches should be always debated with the patient, keeping in mind that right ventricular dysfunction is a marker of illness as well as an indicator of poor survival; thus it should be always anticipated trying to preserve patient survival and quality of life. Therefore, when an outpatient in INTERMACS classes IV–VII begins to show minimal right ventricular dysfunction, it should be immediately referred to an MCS center to warrant the best outcome, avoiding whenever possible to replace a supportable heart.

Many patients with advanced HF have mild to moderate abnormalities of renal function. The serum creatinine concentration may often exceed 2 mg/dl with a creatinine clearance below 50 ml/min, which both have been shown to adversely impact survival after transplantation [63–66]. While renal dysfunction secondary to systemic congestion/impaired perfusion may improve with diuretics or inotropic agents, underlying intrinsic renal disease may represent a significant comorbidity. If cardiac insufficiency is the primary cause of renal dysfunction, it improves after LVAD implant or transplant. During the first 6 months of LVAD support, generally a significant improvement of renal function is observed [67, 68].

When intrinsic renal disease is suspected, a careful nephrologic evaluation is needed, and patients should undergo 24-h urine collection for proteinuria and creatinine clearance, renal ultrasonography looking at kidney size and structure, and possibly evaluation of the renovascular system (sequential SPECT and/or vascular Doppler). Severe renal impairment with GFR lower than 30 ml/min represents a relative contraindication to LVAD, where a high potential recovery should be weighted with an increased hazard of perioperative CRRT, RVF, and mortality. Chronic hemodialysis remains a contraindication for long-term VADs as there are few dialysis centers that accept patients with an LVAD, due to the complex hemodynamic management. Risk of infection is an issue in LVAD patients, and the need of a permanent vascular is an additional risk; few data on safety of peritoneal dialysis exists, but a renal function so profoundly compromised represents a clear risk of complications.

Right HF with hepatic congestion may lead to high transaminase levels, with or without elevated bilirubin, with associated coagulation disorders, but these may be reversible. On the contrary, primary liver diseases and cirrhosis need to be excluded through imaging studies and even a parenchymal biopsy in search of hepatic fibrosis.

Evidence of hepatic synthetic dysfunction with an elevated INR in the absence of warfarin therapy, as disclosed by a high MELD (model for end-stage liver disease) score, is of concern prior to MCS, and attempts should be made to correct this prior to the procedure.

Pulmonary diseases increase the risk of poor outcomes after LVAD implantation, so patients with severe chronic obstructive pulmonary disease or restrictive lung disease are generally excluded. Preoperative mechanical ventilation is one of the driving risk factors leading to the increased risk of mortality of INTERMACS I patients.

Anemia, thrombocytopenia, and coagulopathies have been correlated with poor outcomes following VAD implantation. The use of antiplatelet agents should be discontinued several days prior to surgery whenever possible to prevent perioperative bleeding that could impact perioperative renal dysfunction and right ventricular failure. Heparin-induced thrombocytopenia in patients coming out from ECMO is common. Bleedings are a frequent adverse event following continuous-flow VAD implantation whose pathophysiology is primarily related to angiodysplasia and the development of acquired von Willebrand syndrome. Understanding and management of coagulation before LVAD implantation are pivotal to experience a low rate of complications.

Stroke remains a devastating adverse event following LVAD implantation, and checkup for carotid artery disease is a standard pre-LVAD evaluation in patients at risk for atherosclerotic vascular disease. Severe stenosis may be treated prior to LVAD implantation whenever possible. Interesting data on peripheral circulation in patients undergoing second-generation LVAD implantation disclose that those patients experience a greater loss of pulsatility than those treated with OMM, differently from first-generation devices that showed to restore a normal peripheral circulation. These effects on the peripheral circulation have always to be considered because they may potentially increase the incidence of complications due to the higher shear stress and to the opening of arteriovenous shunts [69].

There are conflicting data about the influence of obesity on transplant outcomes. Obesity may be associated with increased morbidity, complications such as infection, and poor perioperative survival and difficulty identifying an appropriately sized donor heart. Obesity may increase 5-year mortality up to twofold, but the same degree of obesity may not be a contraindication to LVAD implantation and represent an indication to a bridge to candidacy policy to be considered carefully giving the low number of patients experiencing a significant weight-loss and gaining heart transplant candidacy. Infections significantly more frequently occur in the overweight, primarily because of drivelines, which rest within skinfolds. Extremely obese patients had higher rates of device-related infection and re-hospitalization. Bariatric surgery to improve weight loss and reduce potential complications has been suggested, but no sufficient data support this strategy.

Patients with the lowest BMI (<22.9) had the worst prognosis. Cardiac cachexia documented by poor nutrition status and low albumin and total protein concentrations (<3.5 and 6 mg/dl, respectively) is a common and deemed risk factor for patients undergoing MCS, that may be or not related with reversible or not reversible frailty [70], and associated with altered immune function, impaired wound healing, and muscle atrophy, lowering the chance to recovery and increasing the risk of prolonged hospitalizations. Prealbumin and total cholesterol are even more sensitive markers of nutritional status and should be routinely evaluated in all candidates for LVAD therapy.