At the ICU discharge, VAD recipients are usually transferred to a sub-intensive setting where the principal objectives of the physiotherapeutic intervention are improving safe movements and postural passages and improving autonomy of daily activities. Monitoring of the pulmonary function still remains an important aspect. To this end, continuing evaluation of the chest imaging, together with the clinical observation, should be adopted to avoid worsening of the respiratory function if it has been already jeopardized during the ICU stay. Pleural effusion, atelectasis, and lung dysventilation must be also prevented/treated by means of a respiratory program focused on respiratory exercises (◘ Fig. 40.3). Particularly in those showing pulmonary alterations, and that do not require a more intense treatment, respiratory therapy can be continued by means of incentive spirometry exercises, in order to encourage deep breathings and enhance diaphragmatic excursion and chest wall expansion. This practice, during the initial postoperative timeframe, also contributes to obtain the active patient involvement. Respiratory exercises can be scheduled during the day in more than one session; once the patient is trained on the use of the incentive spirometry device, the exercise can be performed autonomously. Furthermore, since the degree of patient’s mobility normally increases during this phase, there is also the need to plan a specific device training activity in order to preserve and guarantee the patient’s safety. Thus, the main goals of the initial rehabilitation treatment are centered on patient’s autonomy and device management: exercise sessions are a good way also to teach the patient how to handle an alarm and a change of batteries. At this stage, active ROM exercises, ambulation, and stair climbing are implemented [27], and flexibility exercises are also encouraged and tailored on patient characteristics. Upper body exercises are delayed until 6–8 weeks to ensure sternal healing. Exercise is stopped in case of subjective intolerance or drop in systolic pressure. Physiotherapist must be trained in emergency procedures in case of device malfunctioning and must be also aware of patient hemodynamic instability and device dislodgement during mobilization [22].

Initial postural passages should be performed and the subject instructed how to change the position safely. These activities, for those patients who have not yet started, should be encouraged beginning with the maintenance of a sitting position at the edge of the bed, progressing toward the sitting position in a chair. Then, the patient can start a more intensive muscular program oriented to develop the ability to gain autonomously an upright position. When postural passages, upright position, and in-bed movements are carried out autonomously, patient is ready to walk with or without a walking frame. Walking, lifting on toes, bending knees, cycling, climbing stairs, and other exercises should be supervised (◘ Figs. 40.4 and 40.5). At this stage, the patient must be confident with the device and must be able to dress it, safely. Pre-LVAD conditions can play a major role during the postoperative rehabilitation. Patients who were already confident with the exercise may be facilitated; hemodynamic stability is needed in order to proceed with the exercise’s intensity [28]. Physical activity can be made also in small groups and the duration can range between 5 and 30 min. Group activities are important as patients need to gain confidence in the equipment and become comfortable with the reactions of those who look at the device. Indeed, facilitation of LVAD acceptance is an important component of rehabilitation. At discharge, it is advisable to set up a follow-up period which comprises sessions of supervised exercise [29]. At the end of the hospitalization, the patient must be able to walk alone and to climb the stairs.

No official guidelines or even recommendations exist to determine the training modality for the patients with a LVAD. It seemed appropriate to build the program on the established recommendations for heart failure since many problems are common to both populations. Therefore, we decided to take care of two aspects of the rehabilitation:

Following the Fick principle, the VO₂ depends from the flow and the peripheral extraction. After LVAD implantation the improvement of VO₂ values is not concomitant with an improvement of the hemodynamic parameters given by the device: this confirms the need of a training oriented to the peripheral muscles [30, 31]. There is a common conclusion in literature to agree on the benefit of revalidation after implantation. With a well-oriented training program and combined exercises, the patients will have a better quality of life; those patients who may still develop an aortic flow during exercise could most probably be able to generate greater efforts.

Physiotherapeutic intervention is planned considering a certain degree of progression along the recovery pathway. Activities should progress from passive ROM to ambulation and resistive ROM exercises [32]. Safety during mobilization is implemented by means of the physiotherapeutic intervention as patient can be instructed how to realize postural passages and how to achieve a safe mobility program considering the safety issues related to the driveline and device management [33]. In the ICU, implanted patient is not able to manage autonomously the device, and safety is completely delegated to the staff.

The LVAD used today has a continuous flow, which does look to be an improvement [34]. The absence of pulse and the measurement of a mean arterial pressure (not always easy) have created the necessity to design the training on the maximal charge obtained during cardiopulmonary exercise testing (CPET). The rehabilitation program must propose a combined training.

The dynamic training is based on a CPET practiced on bicycle, treadmill, or other similar ergometers. The feasibility of a maximal test is well demonstrated in the literature, proved by the respiratory quotients (RQ) reached [35–41]. It is mandatory to obtain this maximal test in order to optimize the structure of the training and also to obtain a unique scale to compare results (intra- and inter-studies).

When this CPET is maximal with a RQ >1.15 and without anomaly, the workload will be fixed to 70–80% of the Watts max. The dynamic training must be completed by muscular reinforcement: the principal muscular groups (arms, legs, back, abdominal) must be trained with specific tools [40, 42–44]. Those specific exercises address qualitative and quantitative muscular changes which are met in the heart failure patients and increased by a prolonged bedrest [45, 46]. Based on the one repetition maximum (RM, charge that the patient can lift one time), exercises are planned at 75% of RM, and performed as two sets of ten repetitions.

Some authors define the limited zone of exercise around 60–70% of the VO2max [47] or 50% of the VO2reserve [45], but they do not explain how the training is adapted afterward.

The interval training has not shown this benefit, in any study, for these types of patients. Anyway, the improvements obtained in patients suffering from cardiac failure, via the peripheral pathway, and the higher workload developed with this type of training, give us the impression that this interval training should be proposed as a complementary modality. For the same reason, a training with work period above the respiratory level 1 seems to be pertinent and feasible [42–48]: cycling, walking, rowing, etc., session of 45–60 min, frequency 3–5 per week, duration 8–10 weeks.

We did not find in the literature any publication and of course no recommendation for the use of neuromuscular electrical stimulation for these patients; this technology is very limited and is more about addressing noncompliant patients. It is the same for stretching and relaxation.

A single study proposes a training of the respiratory muscles [39].

On the contrary, in multiple studies, as for heart failure patients, there is a major interest of a multidisciplinary team in close collaboration with the VAD team [34, 42, 43, 48, 49].

Patient’s progress and rehabilitation outcomes of LVAD recipients admitted to a sub-intensive care setting after ICU discharge should be evaluated. To this end, the Functional Independence Measure (FIM) scale has been described as a suitable tool during in-patient rehabilitation [50–53]. The FIM scale consists of the evaluation of 18 functional and cognitive items: eating, grooming, bathing, upper body dressing, lower body dressing, toileting, bladder, bowel, toilet transfer, bed/chair/wheelchair transfer, tub/shower transfer, walking or wheelchair mobility, stair climbing, comprehension, expression, social interaction, problem solving, and memory. FIM, more than other evaluation tools (i.e., Barthel Index), better intercept the treatment progress thanks to the presence of a number of items that explore various functional areas: the cognitive items also contribute to the evaluation of the social and personal interactions. During in-patient rehabilitation, treatment intensity should be also established in order to guarantee appropriateness and safety of care (◘ Fig. 40.6). In LVAD patients, the use of the Borg scale has been proven effective to check patient’s status while exercising [29, 32, 40]. An exertion of somewhat hard intensity on the Borg scale should be used as a limit to interrupt the exercise [39, 40, 43].

A further evaluation tool consists in the measurement of the daily walking distance which should improve over the rehabilitation treatment: this is a simple, immediate measure which can be obtained in order to verify patient’s progressions [54]. Ambulation distance can be fixed on meters or time and must take into account adverse symptoms [55].

6-min walk test has been shown, at 3.6 months postoperatively, to be a predictor of LVAD mortality, since poor performance (<300 m) on the test is associated with increased mortality [56].

To our best knowledge, there is not yet any specific study in LVAD-supported patients that investigated how to best afford gait and balance problems during cardiac rehabilitation (CR). Some useful suggestions could be derived by a study on aged individuals submitted to CR after coronary bypass heart surgery [57]. In that study, besides other physical and functional evaluations, patients were assessed by a Timed-Up-and-Go (TUG) test, aimed at documenting the time taken to rise from a 43 cm high chair, walk as fast as possible to a mark on the floor 3 m away from the chair, and turn, walk back, and sit down again; [58] a TUG test taking longer than 16 s is considered a predictor of falls in older individuals. In order to improve patients’ balance, in the cited study, the usual aerobic and callisthenic exercises and resistance training aimed at reinforcing legs strength have been complemented by exercises on unstable devices, such as balls and loose platforms, under progressively increasing difficulties. The intervention group showed a highly significant improvement in the TUG test, without any complication linked to the exercises. Similar positive results have been reported also by a study on rehabilitation of chronic pulmonary disease patients, [59] in which balance assessment was performed by the cited TUG test and integrated with four more evaluations: the Berg balance scale (a 14-item scale evaluating activities such as transfers, reaching, turning around, and single-leg stance, graded on a scale ranging from 0 = unable/unsafe to 4 = independent/efficient/safe) [60], the unipedal stance test (patient’s ability to stand on one leg for 45 s) [61], the Tinetti test (a 16-item test divided into two sections: balance (9 items) and gait (7 items), for a total score of 28, where scores <26 indicate high risk of falling) [62], and the activities-specific balance confidence scale (it describes patient’s confidence in performing 16 activities without losing their balance or becoming unsteady, on an 11-point scale) [63]. Although no study evaluating such tests and specific interventions has yet been conducted in LVAD-supported patients, the cited tests and the balance training interventions seem to be possibly useful in the particularly new group of chronic patients represented by LVAD patients. Device education and self-care management must be achieved prior to discharge and are basic conditions for admission to an outpatient rehabilitation program. A multidisciplinary approach including cardiac rehabilitation and prevention staff, contributing staff, and consultant staff is strongly recommended in order to guarantee effectiveness and appropriateness of care in VAD patients [64].

After device implantation, a progressive reduction of left ventricular pressure and volume and a decrease of mean pulmonary artery and wedge pressures are usually observed [37]. Such modifications result in reduction and disappearance of dyspnea. The increased output obtained with the support of the LVAD leads to better perfusion of muscle masses, gradual anatomical and functional muscle fibers reverse modifications, and smoothening of neurohormonal mechanisms, which result in progressive reduction of fatigue. Thus, the majority of patients gradually improve their clinical status, from preimplantation NYHA class IV to class I or II; this improvement is most often noticed after 1 month from implantation, as reported by the HeartMate II Investigators [34]. As soon as the LVAD-supported patient is hemodynamically stable, comprehensive (mainly exercise-based) rehabilitation may begin, aimed at restoring an adequate level of mobility independence [21, 22, 65, 66]. The optimal time to start exercise training is yet to be defined. Some recent studies report beginning of exercise training after 27 ± 15 days [52], after 38 ± 18 days [67], and after 48 ± 38 days [40] when patients are considered clinically stable; this kind of rehabilitation is usually conducted as in-patient rehabilitation. Exercise-based rehabilitation is also advocated in the following period, as long-term ambulatory rehabilitation, with the aim of adequately addressing the challenges that influence patients’ independence and quality of life [68]. What is known nowadays is that exercise training, as part of a multimodal intervention, is safe and effective in LVAD-supported patients, both in those ones waiting for heart transplantation (in which exercise training favorably impacts clinical course and improves post-transplant recovery) and in patients implanted with an LVAD as destination therapy [39, 42, 43, 49, 50, 69, 70]. In LVAD-supported patients, it is well known that regular physical exercise has the potential to progressively reverse – at least partially – the pre-existing negative physical and functional modifications [71]. At the beginning of more intensive rehabilitative activities, some care must be given to the fact that the device could create an obstacle to physical, mainly exercise-based, rehabilitation in a still debilitated patient. Patients supported by LVAD come usually from a long and troublesome history of low cardiac output; one of the main features they often present for weeks or months after beginning of LVAD support is sarcopenia. The reduced muscle masses cause a direct limitation on individual’s capacity to stand and walk with correct balance; the effects are worsened by concomitant presence of autonomic dysfunction that may contribute to orthostatic hypotension and to reduced efficiency of receptor-integrator-effector system with consequent reduced balance/stability control.

During the most common rehabilitative activities practiced in the initial phases of CR, such as walking, treadmill, cyclette, and stepping, there is the risk of accidental falls of the patient or the device. Little is reported in literature about complications linked to such falls. It must be remembered that, in the majority of continuous-flow LVAD (cf-LVAD) models in use, the driveline (from skin to controller) is rather short, and a fall of the device (or a fall of the patient with the driveline caught in the handlebars of the ergometer or treadmill) could lead to disconnection of the LVAD’s external power supply, with awful consequences [72]. In some models of cf-LVADs, spring wire extensions are used between the body cable and the external controller, which could be rather long (up to 2 m or more); patients should be instructed to avoid cable swing while cycling, a situation that could lead to trapping or stretching of the cable by the cycle treadle. In a similar way, patients should be instructed to avoid carrying by hand the bag containing batteries and controller, as the long extension cables could swing near their feet and be accidentally stump. In patients with heart failure, all training intensities have been shown to be effective in improving exercise capacity, while moderate to high intensity aerobic training seems to be more effective to induce reverse left ventricular remodeling [73]. In LVAD-supported patients, we would suggest that exercise training could be conducted at individual anaerobic threshold. A cardiopulmonary exercise test should thus be performed as soon as the patient is able to cycle and repeated by time to time; the effort could be limited to the amount necessary to identify the anaerobic threshold (or slightly more), with greatest attention paid to patient’s symptoms (dyspnea, fatigue, sense of fainting) or clinical signs possibly indicating low peripheral perfusion (paleness, cold sweating, reduction – even small – of mean blood pressure). Performing exercise training at individual’s anaerobic threshold has demonstrated to be safe and effective even in patients rehabilitated rather early, within 2 months from continuous-flow device implantation [74]. It is not yet known if patients could present different levels of effort tolerance as a consequence of different models of implanted cf-LVADs that could present different adaptation of their output to modifications of preload (different ability to accept the contribution given by the systole of the native heart) and afterload (peripheral resistances) linked to physical effort. In the past, in patients supported by pulsatile LVADs, different exercise performances have been reported for pneumatic versus electrically driven devices [75]. As regards cf-LVADs, only one study performed on a mock circulatory system tested and compared performances of an axial-flow and a centrifugal-flow device: in this experimental model, the axial-flow LVAD reached greater maximal output that – if confirmed on clinical ground – could in theory allow better physical performance; it showed anyway also a greater risk of left-ventricle suck-down, which could be counterproductive [76]. Unfortunately, on the clinical ground, no data are yet available about possible different exercise performances allowed by different models of cf-LVADs; in fact, a single study reported the main clinical differences (survival rate, incidence of perioperative bleeding, gastrointestinal bleeding, stroke, renal dysfunction, liver dysfunction, and infection) between two models of cf-LVADs (an axial-flow and a centrifugal-flow device), but did not collect informations about exercise tolerance [77]. During the initial phases of a physical effort, in patients supported by cf-LVADs, mean arterial blood pressure shows an increase whose amount is correlated to the level of METs achieved [55]. Due to the combination of limited maximum output given by the LVAD, poor function of the native heart, and inadequate autonomic regulation of peripheral vascular resistances, blood pressure could progressively reduce at higher intensities of exercise and eventually drop in case of intense effort. However, a substantial problem exists in cf-LVAD-supported patients that blood pressure is not usually measurable by traditional means in clinical practice. A method combining arm-cuff plus Doppler ultrasound identification of humeral artery opening pressure allows to identify “mean” pressure values. It is obvious that during exercise in gymnasium, it is usually difficult to check blood pressure with such method; it is in any case recommended during a cardiopulmonary exercise test, to properly identify the optimal amount of physical activity sustainable by the individual patient. According to treatment protocols for chronic heart failure patients [78], and in an attempt to maximize the unloading of the left ventricle, improve pulmonary resistances, and eventually obtain reversal remodeling of the failing heart [79], LVAD patients are most often treated with a combination of drugs that include angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers associated with selective beta-blockers and aldosterone receptor antagonists drugs. The target for blood pressure control is often put at levels of mean pressure around or below 80 mmHg [80, 81]. By consequence, hypotension with related symptoms is a frequent finding during CR, as well as during normal life activities. Maintenance of an optimal circulating volume is critical to avoid symptomatic hypotension during physical activities, as well as to avoid suction effects induced by the device on interventricular septum, which modify the geometry of the right ventricle and impair its function. Activation of device alarms during physical activity or during changes of patient’s decubitus is most often an indicator of poor “circulating volume,” as it may happen, e.g., due to excessive sweating or important vasodilation after strenuous effort. Thus, patients should be stimulated to regularly introduce small amounts of fluids during and after exercise, in order to compensate for perspiration and avoid relative hypovolemia and consequent hypotension. We would suggest that periodical echocardiographic control of dimensions of left ventricle, position and appearance of interventricular septum, and – if possible – calculation of estimated pulmonary systolic pressure should be performed (all controls that are obviously not possible during performance of rehabilitative physical activities) [82]. Besides being linked to volemia and drugs effects, orthostatic hypotension may be present also as a consequence of the already cited profound cardiac dysautonomia. Autonomic imbalance is present during the first months from the beginning of circulatory support [83] and could progressively improve in the following months [84], reaching an almost normal cardiovascular autonomic homeostasis (and baroreflex activity) after 7–32 months [85]. While in chronic heart failure patients, it is known that exercise training reduces sympathetic outflow and leads to an improvement of baroreflex sensitivity and heart rate variability [86], in cf-LVAD-supported patients, it is not yet known if recovery of a normal autonomic activity could be accelerated by endurance exercise. It is usually not recommended to modify LVAD settings during physical exercise. Even though it should seem logical to try and increase rotational speed in order to parallel the device output with the augmented metabolic demand linked to physical activity, it must be remembered that LVAD output seems to be partially adaptive to changes in activities of daily living, allowing at least low levels of physical engagement [87] and supporting near-normal increments in cardiac output and legs perfusion (with constant cerebral perfusion) during even maximal exercise [88]. While it is true that some experiences report positive functional results with an increase of rotational speed of various models of devices (and, conversely, reduced performance with reduction of their operational speed), especially in cases with lower residual left ventricular function [88–91], the results obtained in these small dimension studies need to be evaluated face to the possible negative effects that the increased rotational pump speed may induce on right ventricle function during sustained submaximal exercise. A possible complication during physical exercise in cf-LVAD-supported patients is the appearance or worsening of atrial and ventricular arrhythmias. During the initial phases of an exercise-based rehabilitation, LVAD patients should ideally be controlled by telemetry ECG monitoring, to check for presence or appearance of major atrial or ventricular arrhythmias; it must be remembered, anyway, that ECG monitoring implies application of chest leads and carrying a transmitter that may be disturbing in a patient who is already carrying the device controller and batteries, connected to an abdominal or retroauricular driveline. Ventricular arrhythmias are rather frequent in LVAD-supported patients; it has been reported that ventricular arrhythmias account for 4.66 events per 100 patients/month in the first 12 months after implant of a cf-LVAD [92]. They are a consequence of the underlying heart disease; they may also be linked to hemodynamic modifications induced by the device and to scars around the outflow cannula. During physical exercise in orthostatic position, an even mild dehydration may reduce the diastolic dimensions of the left ventricle and lead to suction events that may cause ventricular irritation and trigger arrhythmic events [93]. This is another reason to stimulate LVAD patients to introduce repeatedly moderate amounts of fluids during and after exercise. If major or life-threatening arrhythmias occur during rehabilitation activities, they usually do not constitute, anyway, an emergency problem: