Left ventricular assist devices (LVADs) are increasingly used to manage the end-stage heart failure, refractory to maximal medical therapy.
Right ventricular failure (RVF) after LVAD implantation is a frequent complication, associated with significant postoperative morbidity and mortality.1 Although most patients can be maintained with inotropic support, 10–15% will require implantation of right ventricular assist device (RVAD).2
Several techniques have been described to support RVF after LVAD implantation:
- - Central RVAD with cannulation between the right atrium (RA) and the pulmonary artery (PA) to assist the RV with either a temporary continuous flow pump or a more durable pulsatile pump. This strategy allows adequate unloading of the right ventricle and provides adequate transpulmonary flow. However, retrieval of the device requires reopening of the sternum with risk of bleeding or device infection.
- - Percutaneous RVAD such as ABIOMED Impella RP placed from the right femoral vein and drained blood from the inferior vena cava into the PA. This device showed immediate hemodynamic improvements in a small multicenter clinical trial.3 Although the study showed favorable outcomes, the absence of a controlled arm prevents definitive claims with respect to survival outcomes.
- - Peripheral venoarterial extracorporeal life support (ECLS), which provides both cardiac and pulmonary support with simple implantation, no re-sternotomy and less expensive cost. However, by shunting the pulmonary circulation, it reduces LVAD preload and by delivering a retrograde blood flow, it increases LVAD postload with risk of low blood flow through LVAD.
Few years ago, a highly original and simplified approach has been reported,4–6 using extracorporeal membrane oxygenation (ECMO) as a RVAD established between a femoral vein and the PA thanks to a Dacron prosthesis connected at skin level. This approach provides adequate LVAD pre- and postload and is associated with significantly less thromboembolic complications than peripheral venoarterial ECLS.7
We report a series of 44 consecutive continuous flow LVADs, of which 22 patients underwent early simplified temporary RV support, described above.
The aim of this study was to evaluate the outcomes of an early and liberal temporary RVAD (t-RVAD) implantation in a LVAD population.
MATERIALS AND METHODS
We conducted a retrospective and observational study using data from 44 consecutive patients who underwent implantable LVAD insertion by HeartMate II LVAD (Thoratec Corp, Pleasonton, CA) by the same cardiac surgeon in two transplant centers (Pitié Salpêtrière Hospital and Bichat Hospital Paris, France) from February 2012 to September 2014. Written informed consent was obtained before patient inclusion.
Patients were divided into two groups: 22 with isolated LVADs (no t-RVAD group) and 22 with a simplified t-RV support (t-RVAD group).
Patients were freely included in t-RVAD group in case of (Figure 1):
- - severe preoperative RV dysfunction (tricuspid annular plane systolic excursion [TAPSE]) < 15 mm, low tricuspid annular peak systolic velocity < 10 cm/s), or severe tricuspid regurgitation at transthoracic echocardiography;
- - multiorgan dysfunction syndrome (MODS) defined by derangement involving two or more organ systems (hepatic, renal, etc.) or under ECLS in whom evaluation of RV function is extremely difficult;
- - intraoperative RV dysfunction, low LVAD output (<3 L/min), or high inotrope dose (dobutamine > 15 gamma/kg/min).
We did not use invasive hemodynamic data to manage the RV support in our population.
All patients were operated by the same surgeon and the same operative technique. Half of the patients required RV support in addition to the LVAD. Most RV supports were performed during the LVAD implantation. Only two patients had a delayed RVAD implantation, but all t-RVADs were implanted less than 48 hours after the LVAD implantation. No patient with isolated LVAD required the implantation of a RVAD during the follow-up.
For the implantation of RVAD, we chose the simplified technique of temporary RVAD,4–6 which consist in cannulation via a transcutaneous cannula in the femoral vein using Seldinger’s technique. No central RVADs were implanted after LVAD in our population.
Briefly, the venous inflow cannula was placed percutaneously into the RA through a femoral vein. For the outflow cannula, an 8 mm Dacron graft was anastomosed end-to-side to the main PA using a side clamp. The Dacron graft was positioned in front of the LVAD outflow graft and tunneled to its percutaneous exit site under the right costal margin. The arterial outflow graft was then inserted into the Dacron graft and secured using heavy sutures (Figure 2).
Device Management and Weaning
Patients were daily evaluated for RV recovery by collecting noninvasive hemodynamic parameters and blood tests and by performing weaning trials.
The explantation of the t-RVAD was performed in operating theater without chest reopening. The weaning technique was detailed elsewhere.7
This approach allows an easy decannulation without general anesthetic and chest-reopening, but it requires to leave the distal portion of the Dacron graft within the thorax.
The preimplantation transthoracic echocardiograms were also reviewed by several cardiologists specialized in echocardiography who was unaware of the treatment group, but without independent reader, with attention to the following: TAPSE, severity of tricuspid regurgitation, tricuspid valve annulus size, tricuspid annular peak systolic velocity (S wave), and right ventricular (RV) to left ventricular (LV) diameter ratio at end-systole (RV/LV ratio).
RV Risk Score
The Michigan RV risk score was calculated for each patient. This score determined by Matthews et al. assigns points based on four variables, with vasopressor use adding 4 points, creatinine > 2.3 mg/dL adding 3 points, bilirubin > 2 mg/dL adding 2.5 points, and aspartate aminotransferase > 80 IU/dL adding 2 points. A higher score is associated with a greater risk for RVF.8,9
We do not have enough hemodynamic data, like PA pulsatility index or central venous pressure to calculate other main prognostic factors (PAPi, CRITT, etc.).
Data analysis was performed with SPSS 20.0 (SPSS Inc., Chicago, IL).
Continuous data are presented as mean ± standard deviation or as median and compared with the use of a Student’s t-test or two-way analysis of variance, as appropriate.
Categorical data are expressed as percentages and compared using chi-squared tests.
Survival rates between the both groups are presented by the Kaplan–Meier method and p values were calculated with the use of the two-sided log-rank test.
A multivariate Cox proportional hazard regression analysis was also performed, including the following covariates: age, sex, intention to treat, and primary diagnosis. All statistical tests were two-tailed, an alpha value p < 0.05 was considered statistically significant.
Of the 44 patients (38 males and 6 females), 22 (50%) did not require RV support (no t-RVAD group), whereas 22 (50%) needed the addition of a temporary RVAD (t-RVAD group).
Baseline patient characteristics were summarized in Table 1.
In the population, median age was 58.2 ± 12.3 years (range, 21–75 years). The patients of t-RVAD group were significantly younger compared with the other group (54.6 ± 14.2 vs. 61.9 ± 8.9 years; p = 0.046).
A total of 23 patients (52%) had a diagnosis of dilated cardiomyopathy, 16 patients (36%) an ischemic cardiomyopathy (ICM), and 3 patients (7%) an acute myocardial infarction (MI). There was no difference in etiology of the heart disease in both groups.
Nine patients (21%) had a previous cardiac surgery (among five CABG). There were more previous surgeries in the t-RVAD group, but the difference was not significant.
Preimplant conditions and preimplant laboratories’ values were summarized in Table 2.
There were greater levels of urea, alanine aminotransferase, aspartate aminotransferase, total bilirubin, and NT-pro-BNP in patients requiring t-RVAD compared with those requiring only LVAD.
Michigan risk score was significantly higher in the t-RVAD group. Likewise, INTERMACS clinical profile was significantly lower in the t-RVAD group.
Therefore, preimplant conditions were most severe in the t-RVAD group.
The echocardiographic variables were listed in Table 3.
There was no difference in LVEDD, in LVEF, or in moderate to severe mitral regurgitation (MR ≥ 3/4).
TAPSE and S wave were lower in the t-RVAD group and there were more patients with severe tricuspid regurgitation (TR ≥3/4) in the t-RVAD group.
Although echocardiographic variables could be helpful to identify RV failure, none of these was significantly different in both groups.
The median duration of RVAD support was 8.5 days. Among the t-RVAD group, 17 patients (77%) were successfully weaned for RV support, 4 patients (18%) died of bleeding or infectious complications and 1 patient (5%) was transplanted in emergency before RVAD weaning.
Six tricuspid annuloplasties (13.6%) were performed during the LVAD implantation: 2 patients (9.1%) in the no t-RVAD and 4 (18.2%) in the t-RVAD group (p = 0.38). All patients with tricuspid annuloplasties were weaned from the RV support and discharged from the hospital. No patient died and two patients were successfully transplanted.
Four patients were unweanable from their t-RVAD support secondary to persistent right ventricular dysfunction. These four patients with severe preoperative organ dysfunction (high level of creatinine, bilirubin) had an immediate RVAD implantation, but no patient recovered after more than 2 weeks and died during the index hospitalization.
There was no difference in both groups about the lactate release in the first days after the LVAD implantation. In the t-RVAD group, the two patients with delayed RVAD implantation (less than 48 hours after LVAD implantation) had a significantly higher lactate level on postoperative day (POD) 1 compared with patients with an immediate RVAD implantation (2.5 vs. 1.3 mmol/L, p = 0.018), but similar lactate level on POD 5 (1.0 vs. 0.8 mmol/L, p = 0.624). These two patients were successfully weaned from the t-RVAD, but one died 2 weeks after discharge. Mean support duration was lower in the t-RVAD group and there was no difference in terms of pump thrombosis or stroke in both groups during the follow-up.
Finally, survival rate at 6 months (60.4 ± 12 vs. 71.4 ± 9.9%; p = 0.585; Figure 3) and successful bridge to transplant was similar in both groups (46% vs. 46%; p = 1).
Despite severity of preimplant conditions in the t-RVAD group, clinical outcomes did not differ. The postimplantation morbidity of the two groups was listed in Table 4.
Describe the patients receiving a LVAD support who are going to develop RVF is essential to an early initiation of RV support. Indeed, RVF can compromise both LVAD and end-organ function, and these patients had higher mortality, greater risk of bleeding or reoperation, longer hospitalizations, and a higher rate of renal insufficiency (1 year mortality of patients with RVAD was 44% compared with 21% in those who did not require RVAD support [p = 0.007]).9
Unfortunately, it is difficult to predict patients who are needing a RV support after LVAD implantation despite the existence of risk scores for RV failure, including scoring systems with clinical, hemodynamic, laboratory, and echocardiographic parameters.10,11 Indeed, these scoring systems can be cumbersome and some have not been validated in the era of continuous-flow LVADs.
In our study, we found a limited role for preoperative echocardiography. No echocardiographic criterion among TAPSE, tricuspid regurgitation, tricuspid valve annulus size, and tricuspid annular peak systolic velocity was predictive of RVF. This result is matching with other recent studies in whom only RV strain may be a useful preoperative predictor of RV failure in patients undergoing LVAD implantation.12
However, in case of severe tricuspid regurgitation, performing a tricuspid repair during RVAD implantation could improve the success rate of RVAD weaning.
The proportion of LVAD recipients requiring postimplant RV support was slightly higher in our study than in other published reports.1,13 This finding might be explained by an early and liberal RVAD implantation to immediately bring the vicious circle into the end.
Multivariate analysis identified ECLS as one of the independent risk factors for t-RVAD. The explanation for this finding is likely multifactorial, including inadequate decompression of the left heart with ECLS, resulting in lung injury and elevated pulmonary vascular resistance, and ECLS-related complications such as infection affecting end-organ function. However, ECLS is often used before LVAD implantation to stabilize patients in cardiogenic shock or as a bridge to decision. A larger issue might be that the need for preimplantation ECLS reflects the overall severity of both end-organ failure and myocardial failure.
Although patients in the t-RVAD group were more severe, with higher preoperative Michigan scores, our early management led to a mainly comparable outcome with a relatively low overall mortality when compared with survival from historical publications.1,13,14
These results could be explained by an early RVAD implantation, allowing a similar end-organ perfusion in both groups, as evidenced by same lactate release in the first days after the LVAD implantation, insisting on the importance of an early RVAD implantation.
Early initiation of RV support is preferred to a rescue therapy for failing RV function, explaining worse outcomes observed with unplanned RVAD support.15
However, LVADs provide better survival and quality of life than biventricular assist device (BiVAD).10 It is debatable whether the greater mortality with BiVAD use was because patients were sicker and had more organ compromise or whether having two devices resulted into more complications, irrespective of the preoperative status. Having two pumps, instead of one, increases the risk of infection and clot formation requiring more frequent pump changes. The use of a t-RVAD relative to BiVAD allows to decrease these complications and improve outcomes with similar survival rate.16 In consequence, it is important to emphasize that pulsatile extracorporeal BiVAD are not a good option to prevent RVF after LVAD implantation, in so far as it may increase morbidity comparative to t-RVAD.
The main question in the future is to identify patients who are going to develop reversible RV dysfunction needing temporary RV support, those who will not be weanable from t-RVAD, and those who will not require a more durable solution by BiVAD from the start.
Our results should be interpreted in the context of a retrospective study design with a small number of patients and several selection biases. However, use of propensity score is not the solution because the differences in each group were the very essence of this article, based on liberal use of t-RVAD in patients with higher risk or RVF. Some of the preoperative data collected retrospectively were not complete, such as the absence of invasive hemodynamic data, which could introduce some bias. Second, the choice between LVAD and BiVAD was often determined by intraoperative RV performance, and we could not exclude a degree of selection bias in borderline patients with RVF.
This study suggested that RVF occurred in about 50% of LVAD recipients. An early identification of RVF is essential to begin specific therapy, including RV support, and decrease the mortality.
During the LVAD implantation, we chose to early implant a temporary RVAD in case of severe preoperative RV dysfunction, severe tricuspid regurgitation, preoperative MODS or ECLS, and low LVAD output or high inotropes.
This early and liberal use of t-RVAD in case of risk factors of RVF could improve the prognostic in this population by a decrease in overall mortality.
1. Patlolla B, Beygui R, Haddad F. Right-ventricular failure following left ventricle assist device implantation. Curr Opin Cardiol 2013.28: 223233.
2. Deng MC, Edwards LB, Hertz MI, et al.; International Society for Heart and Lung Transplantation: Mechanical circulatory support device database of the International Society for Heart and Lung Transplantation: Third annual report–2005. J Heart Lung Transplant 2005.24: 11821187.
3. Anderson MB, Goldstein J, Milano C, et al. Benefits of a novel percutaneous ventricular assist device for right heart failure: The prospective RECOVER RIGHT study of the Impella RP device. J Heart Lung Transplant 2015.34: 15491560.
4. Strauch JT, Franke UF, Madershahian N, Wahlers T. Right ventricular assist device
implantation—A new transcutaneous approach. Thorac Cardiovasc Surg 2004.52: 378379.
5. De Silva RJ, Soto C, Spratt P. Extra corporeal membrane oxygenation as right heart support following left ventricular assist device
placement: A new cannulation technique. Heart Lung Circ 2012.21: 218220.
6. Lenoir M, Quessard A, N’guyen A, Kirsch M. Simplified temporary right ventricular support after implantation of a left ventricular assist device
. Heart Surg Forum 2013.16: E152E154.
7. Noly PE, Kirsch M, Quessard A, et al. Temporary right ventricular support following left ventricle assist device implantation: A comparison of two techniques. Interact Cardiovasc Thorac Surg 2014.19: 4955.
8. Miller LW, Guglin M. Patient selection for ventricular assist devices: A moving target. J Am Coll Cardiol 2013.61: 12091221.
9. Kormos RL, Teuteberg JJ, Pagani FD, et al.; HeartMate II Clinical Investigators: Right ventricular failure
in patients with the HeartMate II continuous-flow left ventricular assist device
: Incidence, risk factors, and effect on outcomes. J Thorac Cardiovasc Surg 2010.139: 13161324.
10. Matthews JC, Koelling TM, Pagani FD, Aaronson KD. The right ventricular failure
risk score a pre-operative tool for assessing the risk of right ventricular failure
in left ventricular assist device
candidates. J Am Coll Cardiol 2008.51: 21632172.
11. Raina A, Seetha Rammohan HR, Gertz ZM, Rame JE, Woo YJ, Kirkpatrick JN. Postoperative right ventricular failure
after left ventricular assist device
placement is predicted by preoperative echocardiographic structural, hemodynamic, and functional parameters. J Card Fail 2013.19: 1624.
12. Kato TS, Jiang J, Schulze PC, et al. Serial echocardiography using tissue Doppler and speckle tracking imaging to monitor right ventricular failure
before and after left ventricular assist device
surgery. JACC Heart Fail 2013.1: 216222.
13. Lampert BC, Teuteberg JJ. Right ventricular failure
after left ventricular assist devices. J Heart Lung Transplant 2015.34: 11231130.
14. Aissaoui N, Morshuis M, Schoenbrodt M, et al. Temporary right ventricular mechanical circulatory support for the management of right ventricular failure
in critically ill patients. J Thorac Cardiovasc Surg 2013.146: 186191.
15. Takeda K, Naka Y, Yang JA, et al. Outcome of unplanned right ventricular assist device
support for severe right heart failure after implantable left ventricular assist device
insertion. J Heart Lung Transplant 2014.33: 141148.
16. Ochiai Y, McCarthy PM, Smedira NG, et al. Predictors of severe right ventricular failure
after implantable left ventricular assist device
insertion: analysis of 245 patients. Circulation 2002.106(12 Suppl 1): I198I202.