Tricuspid regurgitation (TR) is common in patients with advanced heart failure (HF). Most often, TR in this setting is “functional” in nature, occurring as a result of progressive right ventricular (RV) and tricuspid annular dilatation in response to pulmonary venous hypertension from left-sided heart disease. Given its prevalence, significant TR is encountered in up to half of patients undergoing left ventricular assist device (LVAD) implantation.1 However, significant controversy exists regarding whether the TR should be surgically managed with valvular repair/replacement at the time of LVAD implantation.
Postoperative RV failure is common following LVAD, occurring in up to 44% of patients.2 Patients with RV failure frequently require prolonged intensive care unit stays, implantation of RV assist devices (RVAD) and experience increased mortality.2,3 The presence of significant preoperative TR is associated with increased postoperative RV failure following LVAD.1 In the immediate postoperative period, left ventricular (LV) offloading from the LVAD can contribute to leftward shift of the interventricular septum,4 resulting in changes in RV geometry that may worsen TR and RV function. Furthermore, the regurgitant tricuspid flow can diminish forward flow through the pulmonary circulation thereby reducing LV preload and LVAD flow.
Accordingly, there has been interest in surgically correcting TR at the time of LVAD implantation in an effort to mitigate its potential effects on postoperative RV function. As there are no published guidelines to recommend for or against this practice, the choice has largely been operator-dependent, with some surgeons adopting an aggressive approach to repair. However, others have avoided concomitant tricuspid valve procedures at the time of LVAD to shorten cardiopulmonary bypass time.
To better inform this practice, we aimed to systematically review the available evidence comparing postoperative outcomes in patients treated with LVAD alone versus LVAD with concomitant tricuspid valve surgery (TVS).
The goals of this review were to systematically assess differences in early postoperative outcomes in patients with advanced HF and TR treated with LVAD compared with LVAD + TVS.
We included randomized controlled trials or observational studies, which met the following criteria: 1) included adult human subjects; 2) compared a group of patients treated with LVAD alone versus LVAD + concomitant TVS (tricuspid valve repair or replacement); 3) compared cardiopulmonary bypass time, early mortality, or postoperative RVAD use between groups; and 4) were published in the English language.
On May 10, 2014, we searched SCOPUS, Web of Science, Ovid EMBASE, and Ovid MEDLINE for articles containing the following terms: “LVAD” or “heart assist” or “left ventricular assist” or “heart mate” or “heartware” AND “tricuspid” or “tvr” or “tvpr” AND “procedure” or “repair” or “replace.” We restricted the search to articles published since the year 2000, given the rapid changes in device technology since that time. We did not restrict our criteria to studies solely analyzing patients with continuous-flow LVADs. The full search strategy is included as Supplemental Digital Content (http://links.lww.com/ASAIO/A54).
Two reviewers (SMD and SVD) independently reviewed all abstracts identified using our search strategy. Abstracts that potentially met study criteria were identified and the full-text articles were then reviewed in duplicate to determine the final studies eligible for inclusion.
Outcomes of interest included cardiopulmonary bypass time, need for RVAD postoperatively, duration of inotropic support, post-LVAD hospital length of stay, development of postoperative acute renal failure, and early mortality. We used individual study definitions for postoperative acute renal failure occurring during the same hospitalization, but after LVAD surgery. We defined early mortality as in-hospital or 30 day mortality as reported in the individual studies.
The quality of full-length articles was assessed using the Newcastle Ottawa Scale.
Three authors5–7 reported their continuous outcomes as median (interquartile range), a format that did not allow us to pool the data. We contacted them by email asking them to provide us with means (standard deviation), which only one of the authors5 was able to provide.
We used the I2 statistic, which estimates the percentage of total variation across studies that is due to heterogeneity rather than chance. Suggested thresholds for heterogeneity were used, with I2 values of 25–49%, 50–74%, and ≥75% indicative of low, moderate, and high heterogeneity, respectively. Random effects models were used to pool data. Binary outcomes (RVAD, renal failure, mortality) are presented as risk ratios and continuous variables (cardiopulmonary bypass times) are presented as estimated mean differences. The risk ratios for studies reporting the number of outcomes in each group, rather than an adjusted risk ratio, were calculated and pooled with the adjusted risk ratios reported by Robertson et al.8 We performed sensitivity analyses including 1) restricting to studies that only included patients with continuous-flow LVADs and 2) restricting to studies that accounted for baseline TR severity. Analysis was performed using Stata 13.0 (College Station, Texas) and Review Manager 5.1 (RevMan 5.1®, Nordic Cochrane Center and Copenhagen, Denmark).
Our search strategy identified 277 abstracts. From this group, 28 articles were reviewed, of which 6 met criteria for inclusion in this review. Five studies5–9 were observational in nature and one10 was a post hoc analysis of clinical trial participants; there were no randomized controlled trials on this topic. In total, four studies5–7,9 reflected the experience of a single center. The remaining studies included data from the HeartMate II clinical trials10 and the Society of Thoracic Surgeons (STS) National Database.8 The characteristics of the included studies are shown in Table 1. Only three studies limited their population to patients who would have an indication for TVR (at least moderate TR pre-LVAD).6–8 Three studies only included patients with continuous flow devices,8–10 whereas the remaining three each had a small proportion with pulsatile flow devices. The quality of the included studies as demonstrated by the Newcastle Ottawa Scale ranged from 6 to 9 stars (maximum score = 9 stars). Most studies lost stars for comparability of the patients treated with LVAD + TVS versus LVAD alone, as patients treated with LVAD + TVS, on average, had more preoperative TR and worse RV function (Table 2), and those differences were not accounted for in many of the analyses.
Cardiopulmonary Bypass Time
Four studies including 2,330 patients reported on cardiopulmonary bypass times,5,7–9 though one reported only median times for each group and could not be included in pooled analysis.7 In the remaining three studies, TVS at the time of LVAD prolonged cardiopulmonary bypass time by an average of 31 minutes per patient (Figure 1). Three7–9 out of four studies reported significantly longer bypass times with TVS.
Need for RVAD
All six studies including 3,249 patients reported on the postoperative need for RVAD for RV failure,. In pooled analyses, there was no difference in the need for RVAD in patients treated with LVAD + TVS versus LVAD alone (RR 1.42, 95% CI 0.54–3.76, I2 = 71%, Figure 2). The study by John et al.,10 which analyzed data from the HeartMate II trials, was the sole study to find a significantly higher risk of RVAD insertion in patients treated with LVAD + TVS versus LVAD alone (9.7% vs. 1.9%, p < 0.001). However, this study did not account for differences in baseline risk or degree of TR. In a subgroup analysis including the three studies6–8 that exclusively compared patients with at least moderate TR who underwent LVAD + TVS versus LVAD alone, there was still no difference in postoperative need for RVAD (RR 0.79, 95% CI 0.49–1.28, I2 = 5%, p = 0.34). Results were also similar when limiting the analysis to the three studies that solely included patients with continuous flow devices.8–10
Duration of Inotropic Support
There were three studies representing 249 patients that reported on the postoperative duration of inotropic support.5,6,9 One study reported a significantly shorter duration of inotropic support in patients treated with LVAD + TVS,6 where the median duration after LVAD + TVS was 8 days compared with 10 days after LVAD alone (p = 0.04). The remaining two studies found no difference in inotrope duration in patients treated with LVAD + TVS versus LVAD alone (mean 10.5 vs. 7.8 days, respectively,5 and 8.6 vs. 8 days.9
Acute Renal Failure
Four studies reported on the risk of postoperative acute renal failure in patients treated with LVAD + TVS versus LVAD alone, one of which reported a significantly higher risk in those with concomitant TVS. In pooled analyses including 2,020 patients, there was no difference in the risk of postoperative acute renal failure in patients treated with LVAD + TVS versus LVAD alone (RR 1.07, 95% CI 0.55–2.10, Figure 3). Between-study heterogeneity was moderate (I2 = 67%). The results from Saeed and Robertson showed more acute renal failure in those treated with TVS + LVAD compared with LVAD alone (pooled RR 1.56, 95% CI 1.16–2.11), which differed from the other two studies that demonstrated a trend toward lower risk of acute renal failure in patients treated with TVS (pooled RR 0.61, 95% CI 0.34–1.09). In reviewing the data from Saeed et al., patients with TVS had a rise in creatinine postoperatively (from an average of 1.6 mg/dl preoperatively to 2.3 mg/dl on postoperative day 14) compared with a drop in creatinine in those treated with LVAD alone, but both groups had creatinine levels that were comparable to their preoperative values by postoperative day 30. It is notable that this group reported substantial prolongation in cardiopulmonary bypass times with addition of TVS (median 70 minutes longer than patients treated with LVAD alone). The study from the STS database reported increased risk of new onset renal failure and need for dialysis in patients treated with LVAD + TVS versus LVAD alone, which persisted after accounting for propensity to perform a tricuspid valve procedure (TVP).
Hospital Length of Stay
Four studies including 2,401 patients reported on differences in hospital length of stay in patients treated with LVAD + TVS versus LVAD alone.5,6,8,9 The study pooling data from the STS database reported an adjusted 29% increased risk for prolonged hospital length of stay in patients treated with TVS compared with LVAD alone (RR 1.29, 95% CI 1.16–1.43, p < 0.001).8 One study6 reported a significantly shorter length of stay in patients treated with LVAD + TVS compared with LVAD alone (median 19 vs. 26 days, p = 0.02). Both of these studies, reporting disparate results, restricted their analyses to patients with at least moderate TR. The remaining two studies reported no differences in hospital length of stay in patients treated with LVAD + TVS versus LVAD alone (mean 32 vs. 30 days, respectively5 and 23 vs. 20 days.9
In total, all six studies representing 3,217 patients reported on early mortality, although none reported any difference in patients treated with LVAD + TVS versus LVAD alone. There was no difference in early mortality in patients treated with TVS + LVAD versus LVAD alone (Figure 4). There was low heterogeneity between studies. A subgroup analysis including the three studies that exclusively compared patients with at least moderate TR who underwent LVAD + TVS versus LVAD alone,6–8 there was still no difference in early mortality (RR 0.93, 95% CI 0.67–1.29, p = 0.67). Results were similar when limiting the analysis to the three studies that solely included patients with continuous flow devices.8–10
There were very limited data provided in any of the studies on long-term outcomes. The exception was the report from the HeartMate II trials,10 which reported no difference in 1-year (77% vs. 75%) and 2-year (63% vs. 64%) survival in patients treated with LVAD + TVS versus LVAD alone. Beyond that, three articles6,7,9 provided Kaplan–Meier curves for mortality, though there were very few patients being followed beyond 1 year in any study.
On the basis of our review of the literature and synthesis of available data, although the addition of TVS prolongs cardiopulmonary bypass times, we did not find sufficient evidence to suggest that performance of TVS at the time of LVAD implantation has an effect on early postoperative outcomes. However, most existing studies are inadequately powered and fail to adequately adjust for potential confounders. The two largest studies, analyzing data from the HeartMate II trials10 and STS database,8 found no difference in early mortality in patients treated with TVS at the time of LVAD, though one10 found a higher risk of RVAD and the other8 a higher risk of postoperative renal failure and prolonged length of stay.
TR is extremely common in patients with advanced HF, most often occurring as a result of progressive tricuspid annular enlargement and tricuspid valve leaflet tethering because of RV enlargement. The development of TR often leads to a difficult cycle of progressive RV enlargement, and in turn, worsening regurgitation. In patients undergoing valve surgery for mitral regurgitation, it is known that TR can progress after surgery11 and is associated with worse postoperative outcomes. This recognition has led to the most current guidelines from the European Society of Cardiology recommending that patients who have severe TR undergo surgical intervention at the time of left-sided valve surgery (class I), and such surgery should be considered in patients undergoing left sided heart valve surgery with at least mild TR and a dilated tricuspid annulus (class IIa).12 However, similar guidelines do not exist in patients undergoing LVAD, though the prevalence of at least moderate TR has been estimated to be nearly 50% in recent studies.1,9 Accordingly, whether TR is treated with tricuspid valve repair or replacement at the time of LVAD has largely been based on the preference of the individual surgeon. The tricuspid valve has often been viewed as a “pop-off” valve, and there is concern that fixing the TR may result in progressive RV dysfunction and failure in the early postoperative period. Furthermore, there has been some evidence that effective offloading of the left ventricle with the LVAD can result in improvement in the degree of TR,13,14 though this has been inconsistent across studies with some showing no improvement without surgical repair.15 As severe preoperative TR has been identified as a predictor of adverse postoperative outcomes,16 other centers have taken the approach of surgically repairing the tricuspid valve in patients with significant TR at the time of LVAD implantation in the hopes of improving postoperative outcomes and potentially eliminating the need for future tricuspid procedures.
In this review, we found that patients treated with TVS at the time of LVAD consistently had longer cardiopulmonary bypass times across studies compared with patients who underwent LVAD alone, which is expected. However, we found no differences in postoperative outcomes including use of RVAD, renal dysfunction, and early mortality in patients treated with LVAD+ TVS versus LVAD alone. Only three studies compared the results from patients with at least moderate TR,6–8 but there were still no differences in pooled analyses of outcomes when restricted to these studies. In the remaining studies, patients treated with TVS had more TR and worse baseline RV function, and thus were at greater risk of adverse postoperative outcomes than those who underwent LVAD alone. However, differences in baseline risk were not adjusted for in their analyses. In many cases, the fact that postoperative outcomes were similar despite this difference in risk was interpreted as indication that TVS had a positive effect on postoperative outcomes in these patients. Although there were no randomized trials on the topic to address this question, the sole study that adjusted for propensity to have TVS found a higher risk of renal failure and prolonged hospitalization in patients treated with TVS, though no difference in the need for RVAD or early mortality.8 Analyses of the impact of TVS on postoperative outcomes from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) and the Heartware trials were recently presented inter(national) meetings, though full-text results have not been published. However, neither reported differences in early mortality in their published abstracts.17,18 In addition to the pooled results reported herein, Maltais et al. reported evidence of reverse RV remodeling and improved RV function after LVAD + TVS,9 though data are lacking to inform us as to whether these early changes translate into differences in long-term outcomes. As such, further work is also needed to better evaluate and understand the long-term implications of performing TVS at the time of LVAD. The increasing use of LVAD as destination therapy underscores the importance of examining this issue, as the LVAD remains in situ for the remainder of the patient’s life and the impact of TR on long-term outcomes becomes of great importance.
There are limitations to acknowledge to aid in interpretation of these data. All included studies were observational, and patient selection for TVS may have contributed to results observed. Only three studies restricted their population to patients implanted with continuous flow LVADs, while others included older generation pulsatile flow pumps. Differences in physiology and surgical implantation requirements may contribute to variation in early postoperative outcomes. However, when we restricted our analysis of RVAD and early mortality to those studies using continuous flow devices, we found results to be similar. Some studies included patients who underwent both tricuspid valve repair and replacement. As we did not have access to the individual patient-level data from the various studies, we were unable to assess whether differences in outcomes may occur as a result of the type of TVP employed.
This is the first systematic review and meta-analysis to evaluate the impact of TVS performed at the time of LVAD implantation on cardiopulmonary bypass times and early postoperative outcomes. We found that there are limited observational data available on this topic. Although cardiopulmonary bypass times appear to be longer in those treated with TVS at the time of LVAD implantation, there is insufficient information to draw definitive conclusions on the impact of TVS on early postoperative outcomes. As such, further data are needed in order to inform surgeons as to the best practice for these patients.
The authors thank Patricia Erwin for her assistance with developing a search strategy.
1. Piacentino V 3rd, Williams ML, Depp T, et al. Impact of tricuspid valve regurgitation in patients treated with implantable left ventricular assist devices. Ann Thorac Surg. 2011;91:1342–1346
2. Drakos SG, Janicki L, Horne BD, et al. Risk factors predictive of right ventricular failure after left ventricular assist device implantation. Am J Cardiol. 2010;105:1030–1035
3. 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;139:1316–1324
4. Neragi-Miandoab S, Goldstein D, Bello R, Michler R, D’Alessandro D. Right ventricular dysfunction following continuous flow left ventricular assist device placement in 51 patients: Predicators and outcomes
. J Cardiothorac Surg. 2012;7:60
5. Krishan K, Nair A, Pinney S, Adams DH, Anyanwu AC. Liberal use of tricuspid-valve annuloplasty during left-ventricular assist device implantation. Eur J Cardiothorac Surg. 2012;41:213–217
6. Piacentino V 3rd, Troupes CD, Ganapathi AM, et al. Clinical impact of concomitant tricuspid valve procedures during left ventricular assist device implantation. Ann Thorac Surg. 2011;92:1414–1418
7. Saeed D, Kidambi T, Shalli S, et al. Tricuspid valve repair with left ventricular assist device implantation: Is it warranted? J Heart Lung Transplant. 2011;30:530–535
8. Robertson JO, Grau-Sepulveda MV, Okada S, et al. Concomitant tricuspid valve surgery during implantation of continuous-flow left ventricular assist devices: A Society of Thoracic Surgeons database analysis. J Heart Lung Transplant. 2014;33:609–617
9. Maltais S, Topilsky Y, Tchantchaleishvili V, et al. Surgical treatment of tricuspid valve insufficiency promotes early reverse remodeling in patients with axial-flow left ventricular assist devices. J Thorac Cardiovasc Surg. 2012;143:1370–1376
10. John R, Naka Y, Park SJ, et al. Impact of concurrent surgical valve procedures in patients receiving continuous-flow devices. J Thorac Cardiovasc Surg. 2014;147:581–589
11. Matsunaga A, Duran CM. Progression of tricuspid regurgitation after repaired functional ischemic mitral regurgitation. Circulation. 2005;112(9 Suppl):I453–I457
12. Vahanian A, Alfieri O, Andreotti F, et al. Guidelines on the management of valvular heart disease (version 2012). Eur Heart J. 2012;33:2451–2496
13. Lee S, Kamdar F, Madlon-Kay R, et al. Effects of the HeartMate II continuous-flow left ventricular assist device on right ventricular function. J Heart Lung Transplant. 2010;29:209–215
14. Morgan JA, Paone G, Nemeh HW, et al. Impact of continuous-flow left ventricular assist device support on right ventricular function. J Heart Lung Transplant. 2013;32:398–403
15. Topilsky Y, Oh JK, Atchison FW, et al. Echocardiographic findings in stable outpatients with properly functioning HeartMate II left ventricular assist devices. J Am Soc Echocardiogr. 2011;24:157–169
16. Potapov EV, Stepanenko A, Dandel M, et al. Tricuspid incompetence and geometry of the right ventricle as predictors of right ventricular function after implantation of a left ventricular assist device. J Heart Lung Transplant. 2008;27:1275–1281
17. Milano CA, Pagani FD, Slaughter MS, et al. Clinical outcomes
following implantation of a centrifugal flow left-ventricular assist device and concomitant cardiac valve procedures. Circulation. 2013;128:609
18. Song HK, Mudd JO, Gelow JM, et al. Utility of tricuspid valve repair at the time of left ventricular assist device implantation. J Heart Lung Transplant. 2014;32:537–538