Summary
Ventricular assist devices (VADs) have become an established therapeutic option for patients with end-stage heart failure. The appearance of heart failure in VAD patients seems unexpected. Nevertheless, this phenomenon is not rare. We report six cases of VAD patients with clinical presentation of heart failure at different times after implantation and describe the mechanisms involved. The aetiology of this heart failure, like its clinical presentation, varies and has yet to be identified.
Résumé
L’assistance circulatoire mécanique (ACM) de longue durée est devenue un traitement de l’insuffisance cardiaque terminale. Elle améliore la qualité de vie et la survie de ces patients. L’apparition d’une insuffisance cardiaque chez les patients assistés est inattendue. Pourtant ce phénomène n’est pas rare. Il est important d’identifier cette insuffisance cardiaque et de préciser les mécanismes afin de proposer un traitement adapté. Les objectifs de cette revue sont de décrire les différentes étiologies de l’insuffisance cardiaque sous ACM, leurs présentations cliniques et leurs prises en charge. Cette revue sera illustrée par six cas d’insuffisance cardiaque survenue à des moments différents de l’implantation de l’ACM.
Background
Ventricular assist devices (VADs) have become an established therapeutic option for patients with end-stage heart failure (HF) . VADs support either the left ventricle (LVAD) or both ventricles (BiVAD) . The output of these devices is able to provide a systemic flow close to normal values at rest. Owing to this capability, the occurrence of HF during mechanical support is not expected.
HF is defined as a clinical syndrome characterized by specific symptoms (dyspnoea and fatigue) in the medical history and by signs of fluid retention (pulmonary congestion and peripheral oedema) on physical examination . The usual laboratory tests can help to confirm the diagnosis but their use is limited in VAD patients . The assessment of filling pressure in echocardiography remains difficult because the validity of standard variables is uncertain in severe dilated cardiomyopathy and difficult to prove in LVAD patients when an inflow cannula is implanted in the left ventricle. The use of pulmonary artery catheterization (PAC) can be dangerous in BiVAD patients and is contraindicated in continuous BiVAD and total artificial heart (TAH) patients because it is potentially fatal . We considered acute episodes of left HF, right HF and global HF. The diagnosis was made when dyspnoea with rales and pulmonary congestion was present on X-ray examination (with no indication of infection or acute lung injury) and/or peripheral oedema. The diagnosis was consolidated with increased central venous pressure (CVP) and/or pulmonary capillary wedge pressure (PCWP) when right catheterization or CVP measurements were possible .
If HF occurs, its potential impact on patient outcome can be very significant, in terms of prolonged mechanical ventilation plus its consequences, as well as increased risk of infection multiorgan failure, delayed rehabilitation and longer duration of hospitalization. There are no data in the literature on HF in patients with VADs.
We report six cases in which VAD patients presented HF at different times after implantation and try to describe the different causes of the clinical presentation.
Clinical summaries
HF within the first week post-VAD insertion
Patient 1: HeartWare LVAD – multifactorial HF
A 66-year-old man with ischaemic cardiomyopathy and low ejection fraction had presented New York Heart Association (NYHA) stage IV symptoms since 2009 ( Tables 1 and 2 ). He had chronic renal insufficiency (creatinine 1.8 mg/dL). PAC in June 2009 showed a cardiac index (CI) of 1.6 L/min/m 2 . In November 2010, the patient received a HeartWare LVAD as destination therapy because he developed new cardiogenic shock with progressive decline on inotropic drugs (Interagency Registry for Mechanically Assisted Circulatory Support [INTERMACS] 2). INTERMACS levels are clinical stages that were developed and implemented into the first year of data collection for the INTERMACS ; they are helpful for evaluating the prognosis of patients after VAD implantation. Each level defines clinical status: level 1, critical cardiogenic shock; 2, progressive decline; 3, stable but inotrope dependent; 4, recurrent advanced HF; 5, exertion tolerance; 6, exertion limited; and 7, advanced NYHA III .
| Patient | Age (years) | INTERMACS | Type of VAD | Timing of appearance | Clinical presentation | Aetiology of HF | Outcome |
|---|---|---|---|---|---|---|---|
| 1 | 66 | 2 | LVAD HeartWare | 1 week | Pulmonary oedema | Multifactorial | Discharged |
| 2 | 41 | 1 | BiVAD Thoratec | 1 week | Pulmonary and peripheral oedema | Multifactorial | Discharged |
| 3 | 24 | 3 | LVAD | 2 weeks | Pulmonary oedema | RVF | Discharged |
| 4 | 41 | 1 | BiVAD HeartWare | 1 month | Pulmonary oedema | Pulmonary overflow of RVAD and tamponade | Discharged |
| 5 | 58 | 2 | TAH | 2 months | Pulmonary and peripheral oedema | Multifactorial | Discharged |
| 6 | 72 | – | LVAD VentrAssist | 1 year | Pulmonary and peripheral oedema | Device malfunction | Death |
| Patient | Creatinine | BUN | Plasma protein | ASAT | ALAT | PT | |
|---|---|---|---|---|---|---|---|
| (mg/dL) | (mg/dL) | (g/dL) | (IU/L) | (IU/L) | (%) | ||
| 1 | Before | 1.8 | 62 | 6.1 | 27 | 31 | 83 |
| At the time of HF | 1.9 | 74 | 6.0 | 87 | 15 | 49 | |
| 2 | Before | 2.1 | 205 | 4.5 | 17 | 23 | 42 |
| At the time of HF | 1.8 | 180 | 4.5 | 14 | 2 | 31 | |
| 3 | Before | 0.81 | 13 | 6.2 | 28 | 39 | 70 |
| At the time of HF | 0.7 | 30 | 6.2 | 73 | 64 | 57 | |
| 4 | Before | 1.3 | 79 | 6.6 | 5830 | 2910 | 29 |
| At the time of HF | 0.64 | 32 | 5.9 | 115 | 92 | 79 | |
| 5 | Before | 2.1 | 157 | 5.6 | 293 | 220 | 89 |
| At the time of HF | 1.3 | 100 | 6.1 | 36 | 22 | 29 | |
| 6 | Before | 0.86 | 30 | 7.4 | 33 | 10 | 37 |
| At the time of HF | 3.1 | 116 | 6.7 | 751 | 746 | 8 | |
The PAC measurements on the day of implantation while the patient received inotropic drugs were: CI, 2.26 L/min/m 2 ; CVP, 4 mmHg; and PCWP, 12 mmHg. On the day of surgery the patient received 5470 L of fluids and diuresis was 3350 L. After LVAD implantation, the patient was immediately extubated but ventilation performance was impaired. The estimated flow of the LVAD was 4.9 L/min at 2900 revolutions per minute (rpm). The CVP was 18 mmHg and the PCWP was 27 mmHg. X-ray examination indicated alveolar-interstitial oedema. The patient was treated by non-invasive ventilation and intravenous diuretics. The status of the patient deteriorated further and he needed mechanical ventilation. The patient was finally extubated after 15 days. He was rehabilitated successfully and discharged after a total of 6 weeks of hospital support.
The symptoms could be explained by impaired renal function, relative volume overload and the long period of low cardiac output syndrome.
Patient 2: Biventricular Thoratec VAD – multifactorial HF
A 41-year-old man presented with a 6-month history of dilated cardiomyopathy and a thrombus in the left ventricle. The thrombus was evacuated by ventriculotomy at the end of September 2010 but weaning from the bypass was impossible. The patient received venoarterial extracorporeal membrane oxygenation. Haemodynamic status and echocardiographic findings precluded weaning. The patient received a biventricular Thoratec paracorporeal VAD (BiVAD) because of biventricular dysfunction at the beginning of November 2010; his respiratory status improved and he was extubated after 3 days. Nevertheless, he needed oxygen and developed peripheral oedema. The right ventricular assist device (RVAD) flow was 4.8 L/min and the LVAD flow was 6.6 L/min. Systolic arterial pressures were about 90 mmHg and CVP was 20 mmHg. Renal function was impaired (creatinine 2.1 mg/dL) and plasma protein was decreased (4.5 g/dL). Echocardiography eliminated tamponade and radiography showed alveolar-interstitial oedema. Oedema decreased after 4 days of intravenous high-dose diuretics. Thirty days after the operation, the patient was discharged with BiVAD support.
The clinical presentation could be explained by the long period of low cardiac output syndrome, impaired renal function, poor nutritional status and vasoplegia induced by the two cardiac surgical interventions.
HF within the first month post-VAD insertion
Patient 3: HeartWare LVAD – right ventricular (RV) failure
A 24-year-old man was diagnosed with dilated cardiomyopathy and low ejection fraction (< 25%) present for 6 weeks and of unknown origin. He was hospitalized at the beginning of January 2011 for a first episode of cardiogenic shock with impossibility of weaning from inotropic drugs (INTERMACS 3). PAC (with inotropic drugs) before the implantation revealed: CI, 2.68 L/min/m 2 ; PCWP, 13 mmHg; and CVP, 5 mmHg. The patient was a drug addict and we decided to implant a HeartWare LVAD as a bridge to a decision.
At day 8, the patient presented respiratory insufficiency; CVP was 16 mmHg and the flow of the LVAD was unchanged at around 5 L/min (for a speed of 2800 rpm). Transthoracic echocardiography showed that the left ventricle was not dilated but that the right ventricle was dilated with severe impairment of systolic function.
After reduction of the LVAD flow (2600 rpm) and inotropic drugs, the patient’s status improved. Transthoracic echocardiography showed that the RV diameter was decreased and the left ventricular (LV) diameter was increased.
We deduced RV failure (RVF) worsened by too high a flow speed.
Patient 4: Biventricular HeartWare VAD – pulmonary overflow of the RVAD and tamponade
A 41-year-old woman with dilated cardiomyopathy of unknown aetiology presented in October 2010 with refractory cardiogenic shock (INTERMACS 1). HeartWare BiVAD implantation was performed because of biventricular dysfunction and the primary postoperative course was uneventful.
At day 21, the patient had pulmonary insufficiency without sign of infection. RVAD flow was 5.2 L/min (2400 rpm) and LVAD flow was 6.6 L/min (2600 rpm). The measurement of CVP showed a pressure of 16 mmHg. X-ray examination showed alveolar-interstitial pulmonary oedema. Echocardiography eliminated tamponade.
We concluded that there was pulmonary overflow of the RVAD and the course was favourable with an increase in left pump flow (2900 rpm) and 2 days of intravenous diuretics.
After 4 weeks (day 29), the patient developed dyspnoea with orthopnoea a few hours after withdrawal of the epicardial electrode. The flows decreased: RVAD flow was 3.3 L/min (2400 rpm) and LVAD flow was 3.9 L/min (2900 rpm). CVP increased to 19 mmHg. X-ray investigation showed cardiomegaly ( Fig. 1 ). Tamponade was suspected and confirmed by echocardiography. The patient underwent emergency surgery, the haematoma was evacuated and haemodynamic status stabilized immediately (RVAD flow 4.6 L/min and LVAD flow 5.8 L/min). The patient was rehabilitated.
