Beating Heart Versus Arrested Heart Isolated Tricuspid Valve Surgery Hassina Baraki,1* MD, Shunsuke Saito,1* MD, Ammar Al Ahmad,1 MD, Bernhard Fleischer,1 MD, Axel Haverich,1 MD, and Ingo Kutschka,1 MD Summary We analyzed the long-term results of two surgical techniques (beating versus non-beating) for isolated tricuspid valve (TV) surgery. The long-term results of 92 consecutive patients who underwent isolated TV surgery were analyzed. We compared patients with beating heart (BH) surgery (n = 48) with patients undergoing arrested heart (AH) surgery (n = 44). BH surgery was more frequently chosen in urgent/emergent operations (P = 0.029) and in redo-operations (P < 0.001). Preoperatively, the rates of renal insufficiency (P = 0.002) and EuroSCORE (P = 0.019) were higher in the BH group than in the AH group. There were no differences in perioperative outcomes and 30-day mortality between the groups. However, freedom from reoperation was significantly lower in the BH group compared to the AH group (P = 0.039). We observed a trend towards lower survival rates at 1, 5, and 10 years in the BH group (77%, 54%, and 41%) compared to those of the AH group (86%, 75%, and 72%, P = 0.062). Multivariate Cox hazard model analysis revealed preoperative heart rhythm (P = 0.014, odds ratio [OR] = 2.296) and EuroSCORE (P = 0.022, OR = 1.049) as independent risk factors for mortality after isolated TV surgery. The superiority of BH surgery over AG surgery was not proven. Surgical intervention should be considered early, since patients with elevated EuroSCORES and arrhythmia have significantly higher mortality rates. (Int Heart J 2015; 56: 400-407) Key words: Isolated tricuspid valve surgery, Beating heart surgery, Cardiac surgery, Long-term follow-up
P
atients with tricuspid valve (TV) disease usually remain clinically asymptomatic for a long period of time. When referred to surgery, many of these patients are already in poor clinical condition and in advanced New York Heart Association (NYHA) functional class. The majority of these patients need concomitant procedures (aortic or mitral valve surgery) or have already undergone one or more previous cardiac operations increasing their risk of mortality and morbidity.1-5) Additionally, compared to other heart valve diseases, the incidence of endocarditis caused by infected central venous catheters or pacemaker leads is markedly higher.6,7) Many of these patients have a history of long-term intravenous drug abuse, intracardiac devices, or need hemodialysis.7,8) Although TV repair is known to be the treatment of choice for patients with functional TV regurgitation based on left heart failure,3,9) TV replacement is still the standard treatment for patients with organic TV diseases associated with leaflet destruction.10-12) On the other hand, recent advances in surgical techniques allow for complex leaflet repair of tricuspid valve lesions.13,14) Furthermore, modern reconstructive techniques such as 3-dimensional ring annuloplasty or implantation of artificial cordae are leading to better success rates for
TV repair compared to conventional techniques.6,15) In addition to this progress in surgical techniques, beating heart (BH) surgery for valvular disease has become attractive for many surgeons, since they expect better protection of the heart from ischemia and reperfusion injury.16-25) However, this technique is technically more demanding and may entail drawbacks, such as potential air embolism.19-21) The potential benefits of BH surgery for valve surgery are still controversial.19-21) The purpose of this study was to review our clinical experience of 92 isolated TV procedures and to compare the long-term outcome of patients with BH versus AH TV surgery. We also evaluated the risk factors associated with reoperation and death associated with isolated TV surgery.
Methods From June 1996 to January 2011 a total of 637 patients underwent TV surgery at our institute. Out of this cohort, we retrospectively analysed the data of 92 patients who underwent isolated TV surgery. Data were extracted from the patient charts recorded in the hospital computer database.
From the 1 Department of Cardiac-, Thoracic-, Transplantation and Vascular Surgery, Hannover Medical School, Hannover, Germany. * These authors contributed equally to this study. Address for correspondence: Hassina Baraki, MD, Department of Cardiothoracic Surgery, Otto-von-Guericke University Magdeburg, Leipziger Strasse 44, 39120 Magdeburg, Germany. E-mail: [email protected] Received for publication December 24, 2014. Revised and accepted January 19, 2015. Released in advance online on J-STAGE June 26, 2015. All rights reserved by the International Heart Journal Association. 400
Vol 56 No 4
BEATING VS. ARRESTED HEART TV SURGERY
Our standard surgical approach was a median sternotomy and central cannulation (bicaval venous cannulation and arterial cannulation of the ascending aorta). A right anterolateral thoracotomy was performed exclusively in redo settings. In these cases cardiopulmonary bypass was established by peripheral cannulation of the femoral artery and the femoral and jugular vein. Depending on surgeon preference, the surgical procedure was performed on a beating or arrested heart. Cardiac arrest was achieved by administration of antegrade cold blood cardioplegia. The patients were divided into a beating heart group (BH, n = 48) and an arrested heart (AH, n = 44) group according to whether cross-clamping of the aorta was performed during cardiopulmonary bypass. Tricuspid valve repair was the preferred procedure and TV replacement was only performed if repair was considered to be unfeasible. Data follow-up: Follow-up was performed by contacting the patients, their family members, or their general practitioners by telephone. Patients were asked about their personal activity level, current symptoms, cardiac reoperations, thromboembolic events, and actual anticoagulation management. During follow-up, echocardiography was performed in all survivors and the data were analysed with regard to new or persisting TV dysfunction. Statistical analysis: Continuous variables are expressed as the mean ± standard deviation and categoric variables as frequencies and proportions. All continuous variables were checked for normality using the Shapiro-Wilk test and normal probability plot. For univariate analysis, normally distributed variables were compared using Student’s t-test and non-normally distributed variables were compared with the Mann-Whitney U test. Categoric variables were compared using chi-square analysis or Fisher’s exact test, as appropriate. Risk factor analysis for death within the first 30 postoperative days was performed using univariate and multivariate logistic regression models. Intraindividual differences in NYHA functional class and TV regurgitation grade were compared using the paired t-test. Freedom from TV reoperation and survival were analysed by the Kaplan-Meier method, using the log-rank test for comparisons. Risk factor analyses for TV reoperation and long-term survival were conducted using univariate and multivariate Cox hazard models. Preoperative and operative factors that were found to have values of P < 0.10 in the univariate analyses were included into the multivariate analyses. All P values are 2-sided, and values of P < 0.05 were considered to indicate statistical significance. Statistical analyses were performed with SPSS 19.0 software (SPSS Inc., Chicago, IL, USA).
Results Preoperative characteristics are summarized in Table I. Beating heart operations were chosen more frequently in urgent/emergent (P = 0.029) and in redo operations (P < 0.001). There were more patients with renal insufficiency in the BH group than in the AH group (P = 0.002), and the logistic EuroSCORE was significantly higher in the BH group than in the AH group (P = 0.019). In 42 cases redo surgery was performed 9.0 ± 9.2 years after the following previous cardiac procedures: mitral valve repair/replacement (n = 6), heart transplantation (n = 6), aortic valve surgery (n = 5), mitral and tricuspid surgery (n = 6), correction of congenital
Patient demographics:
401
anomalies (n = 6), tricuspid valve surgery (n = 6), aortic and mitral valve replacement (n = 2), combined aortic, mitral and tricuspid valve surgery (n = 2), pulmonary endarterectomy (n = 1), coronary artery bypass graft (n = 1), and pericardectomy (n = 1). The leading cause of TV insufficiency was endocarditis (n = 29, 31.5%), followed by TV insufficiency secondary to previous left sided heart valve surgery (n = 25, 27.2%), and functional TV insufficiency without the history of left sided valve surgery (n =23, 25.0%). Nine patients (9.8%) required TV redo surgery due to failed De Vega repair after a mean of 8.9 ± 7.7 years. In patients with endocarditis or functional TV insufficiency, AH operations were performed more frequently than BH operations, and vice versa in patients with previous heart operations (P = 0.019) (Table I). Perioperative outcomes: The surgical procedures in both groups are summarized in Table II. Tricuspid valve replacement with a biological prosthesis was the most frequently performed procedure (n = 41, 44.6%), followed by TV repair (n = 34, 37.0%) and TV replacement with a mechanical prosthesis (n = 17, 18.5%). The choice of the TV procedure was independent from the decision for BH or AH (Table II). Early postoperative outcomes are summarized in Table III. Operation time was significantly longer in the BH group than in the AH group (P = 0.033). However, the cardiopulmonary bypass time between the two groups did not differ significantly (P = 0.484). The choice for BH or AH TV surgery had no significant effect on the early postoperative outcome such as intensive care unit stay (P = 0.195) and hospital stay (P = 0.720) (Table I). Four patients (8.3%) in the BH group and 3 patients (6.8%) in the AH group (P = 1.000) required a re-thoracotomy due to postoperative bleeding. Pacemaker implantation due to atrioventricular block grade III became necessary in 6 patients (12.5%) in the BH group compared to 8 patients (18.2%) in the AH group (P = 0.565). Within the first 30 postoperative days 8 patients (16.7%) died in the BH group versus 4 (9.1%) in the AH group (P = 0.360). Early mortality was caused by cardiac failure (n = 6), sepsis (n = 1), and acute respiratory distress syndrome (n = 1) in the BH group, and by cardiac failure (n = 2) and sepsis (n = 2) in the AH group. Risk factors for death within the first 30 postoperative days by univariate logistic regression were age (P = 0.023), NYHA functional class (P < 0.001), heart rhythm (P = 0.038), urgent/emergent operations (P = 0.004), previous multiple heart operation (twice or more, P = 0.024), preoperative renal failure (P = 0.001), dialysis (P = 0.002), peripheral edema (P = 0.021), and EuroSCORE (P < 0.001). None of these variables were detected by multivariate logistic regression as independent risk factors for 30-day mortality. Long-term results: Survivors were followed up for up to 14 years postoperatively. Mean time of follow-up was 4.2 ± 4.0 years in the BH group and 5.5 ± 4.3 and the AH group. Six patients in the BH group and one in the AH group were lost for evaluation. Follow-up was completed in 92.4% of cases. Patient clinical NYHA status was significantly improved in both groups. The mean NYHA class decreased from 3.1 ± 0.7 preoperatively to 2.2 ± 0.9 in the BH group (P < 0.0001), and from 2.8 ± 0.6 to 2.0 ± 0.9 in the AH group (P < 0.0001). During long-term follow-up, echocardiographic examination identified a remaining TV regurgitation grade ≥ II° in 4 patients in the BH group and in 2 patients in the AH group (mean
402
Int Heart J July 2015
BARAKI, ET AL Table I. Patient Preoperative Characteristics Variable
BH (n = 48)
AH (n = 44)
P
Age (years) Female gender (%) NYHA class II (%) III (%) IV (%) Heart rhythm Sinus (%) Atrial fibrillation (%) Paced (%) Urgent/Emergent indication (%) Previous cardiac surgery (%) Twice or more (%) COPD (%) Diabetes mellitus (%) Renal impairment* (%) Hemodialysis (%) Pulmonary hypertension** (%) Ascites (%) Peripheral edema (%) Left ventricular ejection fraction > 50% (%) 30-50% (%) < 30% (%) Right ventricular ejection fraction > 50% (%) 30-50% (%) < 30% (%) Intravenous drug addiction (%) TV insufficiency grade Lateral thoracotomy (%) EuroScore Etiology Endocarditis (%) Secondary to previous left-sided valve operation (%) Functional TV insufficiency (%) Failed DeVega TV repair (%) Others***
53.6 ± 19.5 24 (50.0)
54.3 ± 19.6 24 (54.5)
0.867 0.663 0.206
9 (18.8) 26 (54.2) 13 (27.1)
13 (29.5) 25 (56.8) 6 (13.6)
23 (47.9) 14 (29.2) 11 (22.9) 10 (20.8) 32 (66.7) 9 (18.8) 5 (10.5) 3 (6.3) 21 (43.8) 5 (10.4) 8 (16.7) 11 (22.9) 28 (58.3)
23 (52.3) 13 (29.5) 8 (18.2) 2 (4.5) 10 (22.7) 2 (4.5) 2 (4.5) 1 (2.3) 6 (13.6) 1 (2.3) 13 (29.5) 4 (9.1) 17 (38.6)
38 (79.2) 8 (16.7) 2 (4.2)
36 (81.8) 8 (18.2) 0 (0.0)
25 (52.1) 18 (37.5) 5 (10.4) 9 (18.8) 3.2 ± 1.0 12 (25.0) 16.5 ± 20.8
23 (52.3) 18 (40.9) 3 (6.8) 7 (15.9) 3.4 ± 0.7 8 (18.2) 6.7 ± 1.1
12 (25.0) 17 (35.4) 7 (14.6) 7 (14.6) 5 (10.4)
17 (38.6) 8 (18.2) 16 (36.4) 2 (4.5) 1 (2.3)
0.845
0.029 < 0.001 0.053 0.438 0.618 0.002 0.206 0.142 0.093 0.059 0.390
0.815
0.788 0.450 0.460 0.019 0.012
*
serum creatinine > 200 μmol/L or dialysis dependent, **systolic pulmonary artery pressure > 40 mmHg, ***degeneration of biological TV prosthesis (n = 2), trauma (n = 2), tumor/thrombus (n = 2), BH indicates beating heart; AH, arrested heart; NYHA, New York Heart Association; COPD, chronic obstructive pulmonary disease; EuroSCORE, European System for Cardiac Operative Risk Evaluation; and TV, tricuspid valve.
Table II. Operative Procedures Operation Replacement with biological valve (%) Replacement with mechanical valve (%) Repair (%) De Vega procedure (%) Ring annuloplasty (%) Leaflet repair (%)
BH (n = 48)
AH (n = 44)
P
22 (45.8) 10 (20.8) 16 (33.3) 8 (16.7) 6 (12.5) 2 (4.2)
19 (43.2) 7 (15.9) 18 (40.9) 11 (25.0) 2 (4.5) 5 (11.4)
0.798 0.599 0.452 0.440 0.271 0.253
BH indicates beating heart; and AH, arrested heart.
interval between the TV surgery and follow-up echocardiography: 4.6 ± 3.9 years, the interval was less than 3 years in 48.8%, 3 to 5 years in 17.1%, and more than 5 years in 34.1%). Figure 1 shows the freedom from TV regurgitation grade ≥ II° in the two groups. There were no significant differences between the groups (P = 0.160). All other patients presented with a TV regurgitation grade 0-I°. There were no reports of valve-
related thromboembolic events during follow-up. However, two patients with mechanical valves in the AH group suffered from anticoagulation related haemorrhage; one case of gastrointestinal and another of subarachnoidal bleeding caused by trauma. Both cases could be managed without major subsequent morbidity. In the AH group only one patient had to undergo reopera-
Vol 56 No 4
403
BEATING VS. ARRESTED HEART TV SURGERY Table III. Early Postoperative Outcomes Variable Operation time (minutes) Cardiopulmonary bypass time (minutes) Aortic cross clamp time (minutes) ICU stay (days) Hospital stay (days) Re-thoracotomy due to bleeding (%) Postoperative AV block III (%) Thirty-day mortality* (%)
BH (n = 48)
AH (n = 44)
P
194.0 ± 54.9 89.0 ± 34.2 7.1 ± 10.6 20.7 ± 14.6 4 (8.3) 6 (12.5) 8 (16.7)
170.9 ± 46.9 84.1 ± 31.9 41 ± 20 4.7 ± 6.3 19.7 ± 11.4 3 (6.8) 8 (18.2) 4 (9.1)
0.033 0.484 0.195 0.720 1.000 0.565 0.360
*
Death within the first 30 postoperative days. BH indicates beating heart; AH, arrested heart; ICU, intensive care unit; and AV, atrioventricular.
Figure 1. Kaplan-Meier curve for freedom from TV regurgitation grade ≥ II° during follow-up in the beating heart and arrested heart groups.
tion due to recurrent high grade TV insufficiency 5.7 years after the De Vega procedure, resulting in a freedom from reoperation rate of 100%, 95%, and 95% at 1, 5, and 10 years, respectively, in the AH group. On the other hand, 6 patients needed reoperation in the BH group, resulting in a freedom from reoperation rate of 94%, 87%, and 70% at 1, 5, and 10 years, respectively (Figure 2). The difference in the reoperation rate between the two groups was significant (P = 0.039). The reasons for reoperation in the BH group were: De Vega failure (n = 1), ring dislocation following annuloplasty (n = 1), degeneration of bioprosthesis (n = 2), and TV re-endocarditis caused by continued drug abuse (n = 2). Univariate Cox hazard model analysis revealed none of the preoperative variables or operative procedures, including beating and arrested heart, replacement/repair, mechanical/biological prosthesis, and DeVega/ ring annuloplasty were significant risk factors for TV reoperation. During the follow-up time, 12 patients in the BH group versus 7 patients in the AH group died. Cardiac-related causes of death were significantly more common in the BH group (8/12) compared to the AH group (3/7) (P = 0.039). KaplanMeier survival analysis revealed survival rates of 77%, 54%, and 41% in the BH group and 86%, 75%, and 72% in the AH group at 1, 5 and 10 years, respectively (Figure 3). Survival was slightly better in the AH group, although the difference was not statistically significant (P = 0.062).
Figure 2. Kaplan-Meier curve for freedom from tricuspid valve reoperation during follow-up in the beating heart and arrested heart groups.
Figure 3. Kaplan-Meier curve for survival after tricuspid valve surgery in the beating heart and arrested heart groups.
The following preoperative risk factors were associated with increased mortality in the univariate Cox hazard model: advanced age (P < 0.001), preoperative NYHA functional class (P < 0.001), heart rhythm (P < 0.001), urgent/emergent operations (P = 0.002), redo surgery (P = 0.004), renal impair-
404
Int Heart J July 2015
BARAKI, ET AL
ment (P < 0.001), peripheral edema (P = 0.017), right ventricular ejection fraction (P = 0.041), and EuroSCORE (P < 0.001). The multivariate Cox hazard model revealed preoperative heart rhythm (P = 0.014, odds ratio [OR] 2.296) and EuroSCORE (P = 0.022, OR 1.049) were independent risk factors for mortality after isolated TV surgery (Table IV). Figure 4 shows the survival after TV surgery stratified by the preoperative heart rhythm differentiating in sinus rhythm, atrial fibrillation, and pacemaker dependency. Patients with sinus rhythm had the best survival, while pacemaker dependent patients had the worst survival (P < 0.001).
freedom from re-operation after TV surgery was significantly better in the arrested heart group than in the beating group (Figure 1) and survival was slightly better after the operation in the arrested heart group compared to the beating heart group (Figure 2). In contrast to our findings, Pfannmüller and coworkers have reported outcomes that were as good for BH TV surgery as for AH TV surgery, and recommend BH surgery especially in minimally invasive surgery.23) The results in our study were obviously influenced by the differences in relevant
Discussion We retrospectively analyzed the long-term results of a total of 637 cardiosurgical patients who underwent TV surgery at Hannover Medical School between June 1996 and January 2011. We evaluated the outcome of 92 consecutive patients, who underwent strictly isolated TV surgery, comparing patients operated on either with a beating heart technique or using cardioplegic arrest. To the best of our knowledge, this is the largest study that has focused on isolated TV surgery including echocardiographic follow-up analyses. One of the most important findings of this study is that we were not able to prove that beating heart tricuspid valve surgery is superior to the arrested heart procedure. The beating heart operation had no positive impact on operation time, cardiopulmonary bypass time, intensive care unit stay, hospital stay, or postoperative complications (Table III). Moreover, the
Figure 4. Kaplan-Meier curve for survival after tricuspid valve surgery stratified by patient preoperative heart rhythm.
Table IV. Univariate and Multivariate Cox Hazard Model for Survival After Isolated Tricuspid Valve Surgery Variable
Univariate P
Beating operation Age Female gender NYHA class Heart rhythm Urgent /emergent indication Re-OP COPD Diabetes mellitus Renal impairment* Pulmonary hypertension** Ascites Peripheral edema Left ventricular ejection fraction Right ventricular ejection fraction Intravenous drug addiction TV insufficiency grade Lateral thoracotomy EuroSCORE Etiology of TV disease Biological valve replacement Mechanical valve replacement DeVega annuloplasty Ring annuloplasty Leaflet repair
0.067 < 0.001 0.661 < 0.001 < 0.001 0.002 0.004 0.128 0.836 < 0.001 0.938 0.091 0.017 0.059 0.041 0.078 0.184 0.100 < 0.001 0.565 0.055 0.410 0.916 0.527 0.174
P
Multivariate Odds ratio 95% CI
0.958 0.696
1.029 0.992
0.353 – 3.000 0.954 – 1.032
0.250 0.014 0.940 0.597
1.681 2.296 1.045 1.352
0.694 – 4.074 1.186 – 4.443 0.330 – 3.314 0.442 – 4.139
0.664
1.283
0.446 – 3.692
0.322 0.780 0.197 0.982 0.501
1.766 0.862 1.036 0.999 1.928
0.574 – 5.435 0.305 – 2.440 0.982 – 1.093 0.949 – 1.053 0.285 – 13.029
0.022
1.049
1.007 – 1.093
0.116
2.083
0.834 – 5.204
* serum creatinine > 200 μmol/L or dialysis dependent, **systolic pulmonary artery pressure > 40 mmHg. CI indicates confidence interval; NYHA, New York Heart Association; OP, operation; COPD, chronic obstructive pulmonary disease; and TV, tricuspid valve.
Vol 56 No 4
BEATING VS. ARRESTED HEART TV SURGERY
preoperative conditions between the 2 groups (Table I and Table IV), where the preoperative conditions in the BH group were significantly more severe than in the AH group, and we must keep this in mind when interpreting the results. What we can say from our own results and those of a previous publication 23) is that there was no obvious superiority of the beating heart group over the arrested heart group. Matsumoto and co-workers 16) compared beating versus non-beating heart surgery for left-sided valve surgery in a randomized study of 50 patients and reported shorter cardiopulmonary bypass time (although not statistically significant), markedly reduced catecholamine dosage, and lower peak postoperative creatine kinase isozyme MB as well as troponin T levels. Romano and co-workers 26) also compared 316 patients who underwent redo mitral valve surgery under beating heart conditions with 134 patients who underwent the same procedure under ventricular fibrillatory arrest and reported better outcomes in the beating heart group, such as shorter operative and cardiopulmonary bypass time, lower transfusion requirements, and shorter time to extubation. The most important advantage of beating heart surgery is that there is no need for cardioplegic arrest of the heart, which might cause myocardial hypoxemia, malnutrition, and electrolyte imbalance.16-22) This advantage is especially important in patients with preoperative myocardial hypertrophy and in those with poor ventricular function who require prolonged periods of aortic cross-clamping.19) In the present study, patients who underwent concomitant procedures with left-sided valve surgery were excluded. Therefore, the majority of the included patients did not present with severe left ventricular hypertrophy which is commonly caused by left-sided valve disease. Preoperative left ventricular ejection fraction was > 50% in 80% of the patients in our study, which was markedly better than in previous studies that reported the superiority of the beating heart technique for valve surgery.16,19,26) In addition, the cardiopulmonary bypass time required for our isolated tricuspid valve operations was also shorter than those for left-sided valve operations in previous reports.16,19,26) On the other hand, one of the major disadvantages of beating heart valve surgery is the limited exposure of the surgical field caused by motion of the heart, a filled aortic root, and excess of blood return, all challenging precise surgical manipulation.21) Even though we could not identify the beating heart technique as a risk factor for reoperation in this current study, we strongly suggest that surgeons should carefully and seriously consider beating heart surgery in cases of isolated TV disease, especially considering the fact that our re-operation rate was significantly higher in the beating heart group (Figure 1). Considering the above-mentioned studies for beating heart mitral valve surgery, the advantages of avoiding cardioplegic arrest might become more evident with increasing cross clamp times. Therefore, patients with multi-valve disease or foreseeable complex TV surgery might be the best candidates for a beating heart tricuspid valve procedure. The second most important finding of this study is that the EuroSCORE and the preoperative heart rhythm are both independent predictors of long-term survival after isolated TV surgery. It is already well known that the preoperative EuroSCORE, along with the functional NYHA class, is an important predictor of clinical outcomes after TV surgery.1,2,8) On the contrary, the importance of the preoperative heart rhythm for the postoperative outcome of TV surgery has not yet been de-
405
scribed. Atrial fibrillation is the most common abnormal cardiac rhythm that frequently accompanies valvular heart disease.27,28) Previous reports have shown that preoperative atrial fibrillation is associated with an increased risk of in-hospital mortality and poor early and late outcomes after aortic valve surgery,29) mitral valve surgery,30) and concomitant aortic valve and coronary artery surgery.31) Atrial fibrillation frequently accompanies not only left-sided valve disease but also TV disease. A study from the Mayo Clinic analysed 60 patients with tricuspid regurgitation and reported that 25% of the patients already had atrial fibrillation at admission, and another 14% of asymptomatic patients developed new atrial fibrillation within 10 years.27) However, the impact of preoperative atrial fibrillation on long-term outcome after isolated TV surgery has never been studied in a large series. In the present study, we have clearly demonstrated that the long-term survival after isolated TV surgery was significantly lower in patients with preoperative atrial fibrillation than in patients with sinus rhythm, and even further reduced in pacemaker dependent patients (Figure 3). The tricuspid valve is also named the “forgotten valve” because significant tricuspid regurgitation may be clinically silent for a long period of time, during which progressive right ventricle dilatation and dysfunction may develop. Eventually, the tricuspid regurgitation may be managed with diuretics for symptoms and only considered for surgery when advanced right ventricular dysfunction, or even liver dysfunction or cirrhosis, have developed. With regard to this fact, it is no surprise that TV surgery is the valve operation with the highest risks in terms of morbidity and mortality.32) Although mitral valve surgery has evolved over the past few decades towards progressively earlier intervention, even in selected asymptomatic patients,33) these more aggressive surgical concepts have not been transferred to TV surgery so far. The long-term survival after TV surgery in the present study (Figure 2) was comparable to those reported in other series,1-8,10,34) but we also expect the outcome to improve further by earlier operation, before the patients develop heart failure symptoms and other secondary organ failure.27) In TV surgery, the potential benefit of repair compared to replacement is still a matter of controversy. Guenther, et al reported a significantly better 10-year survival rate of 47% for TV repair versus 37% for replacement in a single centre experience with 416 patients.10) These findings were confirmed by Singh, et al who showed a better 10-year outcome after TV repair (76%) compared to replacement (55%) in a cohort of 250 patients.11) On the other hand, McGrath, et al found no difference in the survival or reoperation rates between repair and replacement in 530 patients with an overall 15-year survival of 20%.1) A recent report from Moraca, et al also failed to show superiority of repair over replacement in a total of 136 propensity-matched patients.35) Importantly, all these previous studies included combined valve procedures. This strongly limits any conclusion about the specific impact of the TV procedure on long-term outcome. In the present study on isolated TV procedures, we found no significant difference in survival between repair and replacement (data not shown). Limitations: We acknowledge some important limitations of our study. First, the study was a retrospective, non-randomized study comparing two groups with relatively small numbers of patients. The decision on surgical procedure (beating or non-
406
BARAKI, ET AL
beating, repair or replacement, and the method of repair) was based on surgeon preference and experience, which resulted in a selection bias and explains the difference in the preoperative data between the 2 groups as shown in Table I. Propensity score matched analysis would have made the results of this study more clear. Unfortunately, we failed to match the patients by propensity score because of the small number of patients and relatively strong bias in patient selection. Moreover, although we found a significant difference in the reoperation free survival rate, the number of patients was too small to detect the risk factors for reoperation. Secondly, we did not analyze more detailed early postoperative data such as catecholamine dosage, peak creatine kinase isozyme MB values, and peak troponin T values to further evaluate the subclinical potential benefits of the beating heart technique. This was due to inconsistent laboratory data acquisition during the observation period. Conclusion: In this study of 92 consecutive patients who underwent isolated TV surgery, no beneficial effect of beating heart surgery was found. Based on the lack of evidence for benefits of beating heart surgery, we recommend using cardioplegic arrest for isolated tricuspid valve surgery, especially in patients without left ventricular hypertrophy or poor heart function, considering the relatively short cross clamp times in the majority of these patients. We also found that the EuroSCORE and the preoperative heart rhythm are significant predictors in terms of long-term survival after isolated TV surgery. Atrial fibrillation and pacemaker dependency are associated with poor survival. Surgical intervention should be considered early in the course of the disease before irreversible heart failure or secondary end-organ problems occur.
Acknowledgments We gratefully acknowledge the assistance of Mrs. Pia Kokemohr for data acquisition and follow-up of patients.
Disclosure Conflict of interest: Baraki and other co-authors have no con-
flict of interest.
References
1. McGrath LB, Gonzalez-Lavin L, Bailey BM, Grunkemeier GL, Fernandez J, Laub GW. Tricuspid valve operations in 530 patients. Twenty-five-year assessment of early and late phase events. J Thorac Cardiovasc Surg 1990; 99: 124-33. 2. Dalrymple-Hay MJ, Leung Y, Ohri SK, et al. Tricuspid valve replacement: bioprostheses are preferable. J Heart Valve Dis 1999; 8: 644-8. 3. McCarthy PM, Bhudia SK, Rajeswaran J, et al. Tricuspid valve repair: durability and risk factors for failure. J Thorac Cardiovasc Surg 2004; 127: 674-85. 4. Murashita T, Okada Y, Kanemitsu H, et al. Fate of functional tricuspid regurgitation after mitral valve repair for degenerative metral regurgitation. Circ J 2013; 77: 2288-94. 5. Izumi C, Miyake M, Takahashi S, et al. Progression of isolated tricuspid regurgitation late after left-sided valve surgery. Circ J 2011; 75: 2902-7.
Int Heart J July 2015
6. Rogers JH, Bolling SF. The tricuspid valve: current perspective and evolving management of tricuspid regurgitation. Circulation 2009; 119: 2718-25. (Review) 7. Capoun R, Thomas M, Caputo M, Asimakopoulos G. Surgical treatment of tricuspid valve endocarditis: a single-centre experience. Perfusion 2010; 25: 169-73. 8. Baraki H, Saito S, Al Ahmad A, et al. Surgical treatment for isolated tricuspid valve endocarditis- long-term follow-up at a single institution. Circ J 2013; 77: 2032-7. 9. Cohn LH. Tricuspid regurgitation secondary to mitral valve disease: when and how to repair. J Card Surg 1994; 9: 237-41. 10. Guenther T, Noebauer C, Mazzitelli D, Busch R, Tassani-Prell P, Lange R. Tricuspid valve surgery: a thirty-year assessment of early and late outcome. Eur J Cardiothorac Surg 2008; 34: 402-9. 11. Singh SK, Tang GH, Maganti MD, et al. Midterm outcomes of tricuspid valve repair versus replacement for organic tricuspid disease. Ann Thorac Surg 2006; 82: 1735-41. 12. Ratnatunga CP, Edwards MB, Dore CJ, Taylor KM. Tricuspid valve replacement: UK Heart Valve Registry mid-term results comparing mechanical and biological prostheses. Ann Thorac Surg 1998; 66: 1940-7. 13. Irwin RB, Luckie M, Khattar RS. Tricuspid regurgitation: contemporary management of a neglected valvular lesion. Postgrad Med J 2010; 86: 648-55. (Review) 14. Sha JM, Yan ZY, Zhu ZY, et al. Early and midterm results of repair of Ebstein’s anomaly with autologous pericardium. Thorac Cardiovasc Surg 2011; 59: 287-92. 15. Filsoufi F, Salzberg SP, Abascal V, Adams DH. Surgical management of functional tricuspid regurgitation with a new remodeling annuloplasty ring. Mt Sinai J Med 2006; 73: 874-9. 16. Matsumoto Y, Watanabe G, Endo M, Sasaki H, Kasashima F, Kosugi I. Efficacy and safety of on-pump beating heart surgery for valvular disease. Ann Thorac Surg 2002; 74: 678-83. 17. Salerno TA, Panos AL, Tian G, Deslauriers R, Calcaterra D, Ricci M. Surgery for cardiac valves and aortic root without cardioplegic arrest (“beating heart”): experience with a new method of myocardial perfusion. J Card Surg 2007; 22: 459-64. 18. Salerno TA, Suarez MR, Panos AL, et al. Efficacy, feasibility, and pitfalls of transseptal approach in beating-heart mitral valve surgery. J Card Surg 2009; 24: 495-8. 19. Ricci M, Macedo FI, Suarez MR, Brown M, Alba J, Salerno TA. Multiple valve surgery with beating heart technique. Ann Thorac Surg 2009; 87: 527-31. 20. Mo A, Lin H, Wen Z, Lu W, Long X, Zhou Y. Efficacy and safety of on-pump beating heart surgery. Ann Thorac Surg 2008; 86: 1914-8. 21. Mo A, Lin H. On-pump beating heart surgery. Heart Lung Circ 2011; 20: 295-304. (Review) 22. Macedo FI, Rodriguez Y, Salerno TA. Myocardial preservation: beating heart techniques. Semin Thorac Cardiovasc Surg 2011; 23: 314-7. 23. Pfannmüller B, Davierwala P, Misfeld M, Borger MA, Garbade J, Mohr FW. Postoperative outcome of isolated tricuspid valve operation using arrested-heart or beating-heart technique. Ann Thorac Surg 2012; 94: 1218-22. 24. Miyairi T, Miura S, Taketani T, et al. Intraoperative assessment for mitral valve competency in a beating heart under retrograde coronary perfusion. Int Heart J 2013; 54: 192-5. 25. Ji Q, Mei Y, Wang X, Ding W. On-pump versus off-pump coronary artery bypass surgery in high-risk patients. Int Heart J 2014; 55: 484-8. 26. Romano MA, Haft JW, Pagani FD, Bolling SF. Beating heart surgery via right thoracotomy for reoperative mitral valve surgery: a safe and effective operative alternative. J Thorac Cardiovasc Surg 2012; 144: 334-9. 27. Messika-Zeitoun D, Thomson H, Bellamy M, et al. Medical and surgical outcome of tricuspid regurgitation caused by flail leaflets. J Thorac Cardiovasc Surg 2004; 128: 296-302. 28. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham study.
Vol 56 No 4
BEATING VS. ARRESTED HEART TV SURGERY
N Engl J Med 1982; 306: 1018-22. 29. Ngaage DL, Schaff HV, Barnes SA, et al. Prognostic implications of preoperative atrial fibrillation in patients undergoing aortic valve replacement: is there an argument for concomitant arrhythmia surgery? Ann Thorac Surg 2006; 82: 1392-9. 30. Wang B, Xu ZY, Han L, Zahng GX, Lu FL, Song ZG. Impact of preoperative atrial fibrillation on mortarity and cardiovascular outcomes of mechanical mitral valve replacement for rheumatic mitral valve disease. Eur J Cardiothorac Surg 2013; 43: 513-9. 31. Saxena A, Dinh D, Dimitriou J, et al. Preoperative atrial fibrillation is an independent risk factor for mid-term mortality after concomitant aortic valve replacement and coronary artery bypass graft surgery. Interact Cardiovasc Thorac Surg 2013; 16: 488-94. 32. Rankin JS, Hammill BG, Ferguson TB Jr, et al. Determinants of operative mortality in valvular heart surgery. J Thorac Cardiovasc Surg 2006; 131: 547-57. 33. Bonow RO, Carabello BA, Chatterjee K, et al; 2006 Writing Com-
407
mittee Members; American College of Cardiology/American Heart Association Task Force. 2008 Focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation 2008; 118: e523-661. 34. Pfannmüller B, Moz M, Misfeld M, et al. Isolated tricuspid valve surgery in patients with previous cardiac surgery. J Thorac Cardiovasc Surg 2012; 146: 841-7. 35. Moraca RJ, Moon MR, Lawton JS, et al. Outcomes of tricuspid valve repair and replacement: a propensity analysis. Ann Thorac Surg 2009; 87: 83-8.
P
atients with tricuspid valve (TV) disease usually remain clinically asymptomatic for a long period of time. When referred to surgery, many of these patients are already in poor clinical condition and in advanced New York Heart Association (NYHA) functional class. The majority of these patients need concomitant procedures (aortic or mitral valve surgery) or have already undergone one or more previous cardiac operations increasing their risk of mortality and morbidity.1-5) Additionally, compared to other heart valve diseases, the incidence of endocarditis caused by infected central venous catheters or pacemaker leads is markedly higher.6,7) Many of these patients have a history of long-term intravenous drug abuse, intracardiac devices, or need hemodialysis.7,8) Although TV repair is known to be the treatment of choice for patients with functional TV regurgitation based on left heart failure,3,9) TV replacement is still the standard treatment for patients with organic TV diseases associated with leaflet destruction.10-12) On the other hand, recent advances in surgical techniques allow for complex leaflet repair of tricuspid valve lesions.13,14) Furthermore, modern reconstructive techniques such as 3-dimensional ring annuloplasty or implantation of artificial cordae are leading to better success rates for
TV repair compared to conventional techniques.6,15) In addition to this progress in surgical techniques, beating heart (BH) surgery for valvular disease has become attractive for many surgeons, since they expect better protection of the heart from ischemia and reperfusion injury.16-25) However, this technique is technically more demanding and may entail drawbacks, such as potential air embolism.19-21) The potential benefits of BH surgery for valve surgery are still controversial.19-21) The purpose of this study was to review our clinical experience of 92 isolated TV procedures and to compare the long-term outcome of patients with BH versus AH TV surgery. We also evaluated the risk factors associated with reoperation and death associated with isolated TV surgery.
Methods From June 1996 to January 2011 a total of 637 patients underwent TV surgery at our institute. Out of this cohort, we retrospectively analysed the data of 92 patients who underwent isolated TV surgery. Data were extracted from the patient charts recorded in the hospital computer database.
From the 1 Department of Cardiac-, Thoracic-, Transplantation and Vascular Surgery, Hannover Medical School, Hannover, Germany. * These authors contributed equally to this study. Address for correspondence: Hassina Baraki, MD, Department of Cardiothoracic Surgery, Otto-von-Guericke University Magdeburg, Leipziger Strasse 44, 39120 Magdeburg, Germany. E-mail: [email protected] Received for publication December 24, 2014. Revised and accepted January 19, 2015. Released in advance online on J-STAGE June 26, 2015. All rights reserved by the International Heart Journal Association. 400
Vol 56 No 4
BEATING VS. ARRESTED HEART TV SURGERY
Our standard surgical approach was a median sternotomy and central cannulation (bicaval venous cannulation and arterial cannulation of the ascending aorta). A right anterolateral thoracotomy was performed exclusively in redo settings. In these cases cardiopulmonary bypass was established by peripheral cannulation of the femoral artery and the femoral and jugular vein. Depending on surgeon preference, the surgical procedure was performed on a beating or arrested heart. Cardiac arrest was achieved by administration of antegrade cold blood cardioplegia. The patients were divided into a beating heart group (BH, n = 48) and an arrested heart (AH, n = 44) group according to whether cross-clamping of the aorta was performed during cardiopulmonary bypass. Tricuspid valve repair was the preferred procedure and TV replacement was only performed if repair was considered to be unfeasible. Data follow-up: Follow-up was performed by contacting the patients, their family members, or their general practitioners by telephone. Patients were asked about their personal activity level, current symptoms, cardiac reoperations, thromboembolic events, and actual anticoagulation management. During follow-up, echocardiography was performed in all survivors and the data were analysed with regard to new or persisting TV dysfunction. Statistical analysis: Continuous variables are expressed as the mean ± standard deviation and categoric variables as frequencies and proportions. All continuous variables were checked for normality using the Shapiro-Wilk test and normal probability plot. For univariate analysis, normally distributed variables were compared using Student’s t-test and non-normally distributed variables were compared with the Mann-Whitney U test. Categoric variables were compared using chi-square analysis or Fisher’s exact test, as appropriate. Risk factor analysis for death within the first 30 postoperative days was performed using univariate and multivariate logistic regression models. Intraindividual differences in NYHA functional class and TV regurgitation grade were compared using the paired t-test. Freedom from TV reoperation and survival were analysed by the Kaplan-Meier method, using the log-rank test for comparisons. Risk factor analyses for TV reoperation and long-term survival were conducted using univariate and multivariate Cox hazard models. Preoperative and operative factors that were found to have values of P < 0.10 in the univariate analyses were included into the multivariate analyses. All P values are 2-sided, and values of P < 0.05 were considered to indicate statistical significance. Statistical analyses were performed with SPSS 19.0 software (SPSS Inc., Chicago, IL, USA).
Results Preoperative characteristics are summarized in Table I. Beating heart operations were chosen more frequently in urgent/emergent (P = 0.029) and in redo operations (P < 0.001). There were more patients with renal insufficiency in the BH group than in the AH group (P = 0.002), and the logistic EuroSCORE was significantly higher in the BH group than in the AH group (P = 0.019). In 42 cases redo surgery was performed 9.0 ± 9.2 years after the following previous cardiac procedures: mitral valve repair/replacement (n = 6), heart transplantation (n = 6), aortic valve surgery (n = 5), mitral and tricuspid surgery (n = 6), correction of congenital
Patient demographics:
401
anomalies (n = 6), tricuspid valve surgery (n = 6), aortic and mitral valve replacement (n = 2), combined aortic, mitral and tricuspid valve surgery (n = 2), pulmonary endarterectomy (n = 1), coronary artery bypass graft (n = 1), and pericardectomy (n = 1). The leading cause of TV insufficiency was endocarditis (n = 29, 31.5%), followed by TV insufficiency secondary to previous left sided heart valve surgery (n = 25, 27.2%), and functional TV insufficiency without the history of left sided valve surgery (n =23, 25.0%). Nine patients (9.8%) required TV redo surgery due to failed De Vega repair after a mean of 8.9 ± 7.7 years. In patients with endocarditis or functional TV insufficiency, AH operations were performed more frequently than BH operations, and vice versa in patients with previous heart operations (P = 0.019) (Table I). Perioperative outcomes: The surgical procedures in both groups are summarized in Table II. Tricuspid valve replacement with a biological prosthesis was the most frequently performed procedure (n = 41, 44.6%), followed by TV repair (n = 34, 37.0%) and TV replacement with a mechanical prosthesis (n = 17, 18.5%). The choice of the TV procedure was independent from the decision for BH or AH (Table II). Early postoperative outcomes are summarized in Table III. Operation time was significantly longer in the BH group than in the AH group (P = 0.033). However, the cardiopulmonary bypass time between the two groups did not differ significantly (P = 0.484). The choice for BH or AH TV surgery had no significant effect on the early postoperative outcome such as intensive care unit stay (P = 0.195) and hospital stay (P = 0.720) (Table I). Four patients (8.3%) in the BH group and 3 patients (6.8%) in the AH group (P = 1.000) required a re-thoracotomy due to postoperative bleeding. Pacemaker implantation due to atrioventricular block grade III became necessary in 6 patients (12.5%) in the BH group compared to 8 patients (18.2%) in the AH group (P = 0.565). Within the first 30 postoperative days 8 patients (16.7%) died in the BH group versus 4 (9.1%) in the AH group (P = 0.360). Early mortality was caused by cardiac failure (n = 6), sepsis (n = 1), and acute respiratory distress syndrome (n = 1) in the BH group, and by cardiac failure (n = 2) and sepsis (n = 2) in the AH group. Risk factors for death within the first 30 postoperative days by univariate logistic regression were age (P = 0.023), NYHA functional class (P < 0.001), heart rhythm (P = 0.038), urgent/emergent operations (P = 0.004), previous multiple heart operation (twice or more, P = 0.024), preoperative renal failure (P = 0.001), dialysis (P = 0.002), peripheral edema (P = 0.021), and EuroSCORE (P < 0.001). None of these variables were detected by multivariate logistic regression as independent risk factors for 30-day mortality. Long-term results: Survivors were followed up for up to 14 years postoperatively. Mean time of follow-up was 4.2 ± 4.0 years in the BH group and 5.5 ± 4.3 and the AH group. Six patients in the BH group and one in the AH group were lost for evaluation. Follow-up was completed in 92.4% of cases. Patient clinical NYHA status was significantly improved in both groups. The mean NYHA class decreased from 3.1 ± 0.7 preoperatively to 2.2 ± 0.9 in the BH group (P < 0.0001), and from 2.8 ± 0.6 to 2.0 ± 0.9 in the AH group (P < 0.0001). During long-term follow-up, echocardiographic examination identified a remaining TV regurgitation grade ≥ II° in 4 patients in the BH group and in 2 patients in the AH group (mean
402
Int Heart J July 2015
BARAKI, ET AL Table I. Patient Preoperative Characteristics Variable
BH (n = 48)
AH (n = 44)
P
Age (years) Female gender (%) NYHA class II (%) III (%) IV (%) Heart rhythm Sinus (%) Atrial fibrillation (%) Paced (%) Urgent/Emergent indication (%) Previous cardiac surgery (%) Twice or more (%) COPD (%) Diabetes mellitus (%) Renal impairment* (%) Hemodialysis (%) Pulmonary hypertension** (%) Ascites (%) Peripheral edema (%) Left ventricular ejection fraction > 50% (%) 30-50% (%) < 30% (%) Right ventricular ejection fraction > 50% (%) 30-50% (%) < 30% (%) Intravenous drug addiction (%) TV insufficiency grade Lateral thoracotomy (%) EuroScore Etiology Endocarditis (%) Secondary to previous left-sided valve operation (%) Functional TV insufficiency (%) Failed DeVega TV repair (%) Others***
53.6 ± 19.5 24 (50.0)
54.3 ± 19.6 24 (54.5)
0.867 0.663 0.206
9 (18.8) 26 (54.2) 13 (27.1)
13 (29.5) 25 (56.8) 6 (13.6)
23 (47.9) 14 (29.2) 11 (22.9) 10 (20.8) 32 (66.7) 9 (18.8) 5 (10.5) 3 (6.3) 21 (43.8) 5 (10.4) 8 (16.7) 11 (22.9) 28 (58.3)
23 (52.3) 13 (29.5) 8 (18.2) 2 (4.5) 10 (22.7) 2 (4.5) 2 (4.5) 1 (2.3) 6 (13.6) 1 (2.3) 13 (29.5) 4 (9.1) 17 (38.6)
38 (79.2) 8 (16.7) 2 (4.2)
36 (81.8) 8 (18.2) 0 (0.0)
25 (52.1) 18 (37.5) 5 (10.4) 9 (18.8) 3.2 ± 1.0 12 (25.0) 16.5 ± 20.8
23 (52.3) 18 (40.9) 3 (6.8) 7 (15.9) 3.4 ± 0.7 8 (18.2) 6.7 ± 1.1
12 (25.0) 17 (35.4) 7 (14.6) 7 (14.6) 5 (10.4)
17 (38.6) 8 (18.2) 16 (36.4) 2 (4.5) 1 (2.3)
0.845
0.029 < 0.001 0.053 0.438 0.618 0.002 0.206 0.142 0.093 0.059 0.390
0.815
0.788 0.450 0.460 0.019 0.012
*
serum creatinine > 200 μmol/L or dialysis dependent, **systolic pulmonary artery pressure > 40 mmHg, ***degeneration of biological TV prosthesis (n = 2), trauma (n = 2), tumor/thrombus (n = 2), BH indicates beating heart; AH, arrested heart; NYHA, New York Heart Association; COPD, chronic obstructive pulmonary disease; EuroSCORE, European System for Cardiac Operative Risk Evaluation; and TV, tricuspid valve.
Table II. Operative Procedures Operation Replacement with biological valve (%) Replacement with mechanical valve (%) Repair (%) De Vega procedure (%) Ring annuloplasty (%) Leaflet repair (%)
BH (n = 48)
AH (n = 44)
P
22 (45.8) 10 (20.8) 16 (33.3) 8 (16.7) 6 (12.5) 2 (4.2)
19 (43.2) 7 (15.9) 18 (40.9) 11 (25.0) 2 (4.5) 5 (11.4)
0.798 0.599 0.452 0.440 0.271 0.253
BH indicates beating heart; and AH, arrested heart.
interval between the TV surgery and follow-up echocardiography: 4.6 ± 3.9 years, the interval was less than 3 years in 48.8%, 3 to 5 years in 17.1%, and more than 5 years in 34.1%). Figure 1 shows the freedom from TV regurgitation grade ≥ II° in the two groups. There were no significant differences between the groups (P = 0.160). All other patients presented with a TV regurgitation grade 0-I°. There were no reports of valve-
related thromboembolic events during follow-up. However, two patients with mechanical valves in the AH group suffered from anticoagulation related haemorrhage; one case of gastrointestinal and another of subarachnoidal bleeding caused by trauma. Both cases could be managed without major subsequent morbidity. In the AH group only one patient had to undergo reopera-
Vol 56 No 4
403
BEATING VS. ARRESTED HEART TV SURGERY Table III. Early Postoperative Outcomes Variable Operation time (minutes) Cardiopulmonary bypass time (minutes) Aortic cross clamp time (minutes) ICU stay (days) Hospital stay (days) Re-thoracotomy due to bleeding (%) Postoperative AV block III (%) Thirty-day mortality* (%)
BH (n = 48)
AH (n = 44)
P
194.0 ± 54.9 89.0 ± 34.2 7.1 ± 10.6 20.7 ± 14.6 4 (8.3) 6 (12.5) 8 (16.7)
170.9 ± 46.9 84.1 ± 31.9 41 ± 20 4.7 ± 6.3 19.7 ± 11.4 3 (6.8) 8 (18.2) 4 (9.1)
0.033 0.484 0.195 0.720 1.000 0.565 0.360
*
Death within the first 30 postoperative days. BH indicates beating heart; AH, arrested heart; ICU, intensive care unit; and AV, atrioventricular.
Figure 1. Kaplan-Meier curve for freedom from TV regurgitation grade ≥ II° during follow-up in the beating heart and arrested heart groups.
tion due to recurrent high grade TV insufficiency 5.7 years after the De Vega procedure, resulting in a freedom from reoperation rate of 100%, 95%, and 95% at 1, 5, and 10 years, respectively, in the AH group. On the other hand, 6 patients needed reoperation in the BH group, resulting in a freedom from reoperation rate of 94%, 87%, and 70% at 1, 5, and 10 years, respectively (Figure 2). The difference in the reoperation rate between the two groups was significant (P = 0.039). The reasons for reoperation in the BH group were: De Vega failure (n = 1), ring dislocation following annuloplasty (n = 1), degeneration of bioprosthesis (n = 2), and TV re-endocarditis caused by continued drug abuse (n = 2). Univariate Cox hazard model analysis revealed none of the preoperative variables or operative procedures, including beating and arrested heart, replacement/repair, mechanical/biological prosthesis, and DeVega/ ring annuloplasty were significant risk factors for TV reoperation. During the follow-up time, 12 patients in the BH group versus 7 patients in the AH group died. Cardiac-related causes of death were significantly more common in the BH group (8/12) compared to the AH group (3/7) (P = 0.039). KaplanMeier survival analysis revealed survival rates of 77%, 54%, and 41% in the BH group and 86%, 75%, and 72% in the AH group at 1, 5 and 10 years, respectively (Figure 3). Survival was slightly better in the AH group, although the difference was not statistically significant (P = 0.062).
Figure 2. Kaplan-Meier curve for freedom from tricuspid valve reoperation during follow-up in the beating heart and arrested heart groups.
Figure 3. Kaplan-Meier curve for survival after tricuspid valve surgery in the beating heart and arrested heart groups.
The following preoperative risk factors were associated with increased mortality in the univariate Cox hazard model: advanced age (P < 0.001), preoperative NYHA functional class (P < 0.001), heart rhythm (P < 0.001), urgent/emergent operations (P = 0.002), redo surgery (P = 0.004), renal impair-
404
Int Heart J July 2015
BARAKI, ET AL
ment (P < 0.001), peripheral edema (P = 0.017), right ventricular ejection fraction (P = 0.041), and EuroSCORE (P < 0.001). The multivariate Cox hazard model revealed preoperative heart rhythm (P = 0.014, odds ratio [OR] 2.296) and EuroSCORE (P = 0.022, OR 1.049) were independent risk factors for mortality after isolated TV surgery (Table IV). Figure 4 shows the survival after TV surgery stratified by the preoperative heart rhythm differentiating in sinus rhythm, atrial fibrillation, and pacemaker dependency. Patients with sinus rhythm had the best survival, while pacemaker dependent patients had the worst survival (P < 0.001).
freedom from re-operation after TV surgery was significantly better in the arrested heart group than in the beating group (Figure 1) and survival was slightly better after the operation in the arrested heart group compared to the beating heart group (Figure 2). In contrast to our findings, Pfannmüller and coworkers have reported outcomes that were as good for BH TV surgery as for AH TV surgery, and recommend BH surgery especially in minimally invasive surgery.23) The results in our study were obviously influenced by the differences in relevant
Discussion We retrospectively analyzed the long-term results of a total of 637 cardiosurgical patients who underwent TV surgery at Hannover Medical School between June 1996 and January 2011. We evaluated the outcome of 92 consecutive patients, who underwent strictly isolated TV surgery, comparing patients operated on either with a beating heart technique or using cardioplegic arrest. To the best of our knowledge, this is the largest study that has focused on isolated TV surgery including echocardiographic follow-up analyses. One of the most important findings of this study is that we were not able to prove that beating heart tricuspid valve surgery is superior to the arrested heart procedure. The beating heart operation had no positive impact on operation time, cardiopulmonary bypass time, intensive care unit stay, hospital stay, or postoperative complications (Table III). Moreover, the
Figure 4. Kaplan-Meier curve for survival after tricuspid valve surgery stratified by patient preoperative heart rhythm.
Table IV. Univariate and Multivariate Cox Hazard Model for Survival After Isolated Tricuspid Valve Surgery Variable
Univariate P
Beating operation Age Female gender NYHA class Heart rhythm Urgent /emergent indication Re-OP COPD Diabetes mellitus Renal impairment* Pulmonary hypertension** Ascites Peripheral edema Left ventricular ejection fraction Right ventricular ejection fraction Intravenous drug addiction TV insufficiency grade Lateral thoracotomy EuroSCORE Etiology of TV disease Biological valve replacement Mechanical valve replacement DeVega annuloplasty Ring annuloplasty Leaflet repair
0.067 < 0.001 0.661 < 0.001 < 0.001 0.002 0.004 0.128 0.836 < 0.001 0.938 0.091 0.017 0.059 0.041 0.078 0.184 0.100 < 0.001 0.565 0.055 0.410 0.916 0.527 0.174
P
Multivariate Odds ratio 95% CI
0.958 0.696
1.029 0.992
0.353 – 3.000 0.954 – 1.032
0.250 0.014 0.940 0.597
1.681 2.296 1.045 1.352
0.694 – 4.074 1.186 – 4.443 0.330 – 3.314 0.442 – 4.139
0.664
1.283
0.446 – 3.692
0.322 0.780 0.197 0.982 0.501
1.766 0.862 1.036 0.999 1.928
0.574 – 5.435 0.305 – 2.440 0.982 – 1.093 0.949 – 1.053 0.285 – 13.029
0.022
1.049
1.007 – 1.093
0.116
2.083
0.834 – 5.204
* serum creatinine > 200 μmol/L or dialysis dependent, **systolic pulmonary artery pressure > 40 mmHg. CI indicates confidence interval; NYHA, New York Heart Association; OP, operation; COPD, chronic obstructive pulmonary disease; and TV, tricuspid valve.
Vol 56 No 4
BEATING VS. ARRESTED HEART TV SURGERY
preoperative conditions between the 2 groups (Table I and Table IV), where the preoperative conditions in the BH group were significantly more severe than in the AH group, and we must keep this in mind when interpreting the results. What we can say from our own results and those of a previous publication 23) is that there was no obvious superiority of the beating heart group over the arrested heart group. Matsumoto and co-workers 16) compared beating versus non-beating heart surgery for left-sided valve surgery in a randomized study of 50 patients and reported shorter cardiopulmonary bypass time (although not statistically significant), markedly reduced catecholamine dosage, and lower peak postoperative creatine kinase isozyme MB as well as troponin T levels. Romano and co-workers 26) also compared 316 patients who underwent redo mitral valve surgery under beating heart conditions with 134 patients who underwent the same procedure under ventricular fibrillatory arrest and reported better outcomes in the beating heart group, such as shorter operative and cardiopulmonary bypass time, lower transfusion requirements, and shorter time to extubation. The most important advantage of beating heart surgery is that there is no need for cardioplegic arrest of the heart, which might cause myocardial hypoxemia, malnutrition, and electrolyte imbalance.16-22) This advantage is especially important in patients with preoperative myocardial hypertrophy and in those with poor ventricular function who require prolonged periods of aortic cross-clamping.19) In the present study, patients who underwent concomitant procedures with left-sided valve surgery were excluded. Therefore, the majority of the included patients did not present with severe left ventricular hypertrophy which is commonly caused by left-sided valve disease. Preoperative left ventricular ejection fraction was > 50% in 80% of the patients in our study, which was markedly better than in previous studies that reported the superiority of the beating heart technique for valve surgery.16,19,26) In addition, the cardiopulmonary bypass time required for our isolated tricuspid valve operations was also shorter than those for left-sided valve operations in previous reports.16,19,26) On the other hand, one of the major disadvantages of beating heart valve surgery is the limited exposure of the surgical field caused by motion of the heart, a filled aortic root, and excess of blood return, all challenging precise surgical manipulation.21) Even though we could not identify the beating heart technique as a risk factor for reoperation in this current study, we strongly suggest that surgeons should carefully and seriously consider beating heart surgery in cases of isolated TV disease, especially considering the fact that our re-operation rate was significantly higher in the beating heart group (Figure 1). Considering the above-mentioned studies for beating heart mitral valve surgery, the advantages of avoiding cardioplegic arrest might become more evident with increasing cross clamp times. Therefore, patients with multi-valve disease or foreseeable complex TV surgery might be the best candidates for a beating heart tricuspid valve procedure. The second most important finding of this study is that the EuroSCORE and the preoperative heart rhythm are both independent predictors of long-term survival after isolated TV surgery. It is already well known that the preoperative EuroSCORE, along with the functional NYHA class, is an important predictor of clinical outcomes after TV surgery.1,2,8) On the contrary, the importance of the preoperative heart rhythm for the postoperative outcome of TV surgery has not yet been de-
405
scribed. Atrial fibrillation is the most common abnormal cardiac rhythm that frequently accompanies valvular heart disease.27,28) Previous reports have shown that preoperative atrial fibrillation is associated with an increased risk of in-hospital mortality and poor early and late outcomes after aortic valve surgery,29) mitral valve surgery,30) and concomitant aortic valve and coronary artery surgery.31) Atrial fibrillation frequently accompanies not only left-sided valve disease but also TV disease. A study from the Mayo Clinic analysed 60 patients with tricuspid regurgitation and reported that 25% of the patients already had atrial fibrillation at admission, and another 14% of asymptomatic patients developed new atrial fibrillation within 10 years.27) However, the impact of preoperative atrial fibrillation on long-term outcome after isolated TV surgery has never been studied in a large series. In the present study, we have clearly demonstrated that the long-term survival after isolated TV surgery was significantly lower in patients with preoperative atrial fibrillation than in patients with sinus rhythm, and even further reduced in pacemaker dependent patients (Figure 3). The tricuspid valve is also named the “forgotten valve” because significant tricuspid regurgitation may be clinically silent for a long period of time, during which progressive right ventricle dilatation and dysfunction may develop. Eventually, the tricuspid regurgitation may be managed with diuretics for symptoms and only considered for surgery when advanced right ventricular dysfunction, or even liver dysfunction or cirrhosis, have developed. With regard to this fact, it is no surprise that TV surgery is the valve operation with the highest risks in terms of morbidity and mortality.32) Although mitral valve surgery has evolved over the past few decades towards progressively earlier intervention, even in selected asymptomatic patients,33) these more aggressive surgical concepts have not been transferred to TV surgery so far. The long-term survival after TV surgery in the present study (Figure 2) was comparable to those reported in other series,1-8,10,34) but we also expect the outcome to improve further by earlier operation, before the patients develop heart failure symptoms and other secondary organ failure.27) In TV surgery, the potential benefit of repair compared to replacement is still a matter of controversy. Guenther, et al reported a significantly better 10-year survival rate of 47% for TV repair versus 37% for replacement in a single centre experience with 416 patients.10) These findings were confirmed by Singh, et al who showed a better 10-year outcome after TV repair (76%) compared to replacement (55%) in a cohort of 250 patients.11) On the other hand, McGrath, et al found no difference in the survival or reoperation rates between repair and replacement in 530 patients with an overall 15-year survival of 20%.1) A recent report from Moraca, et al also failed to show superiority of repair over replacement in a total of 136 propensity-matched patients.35) Importantly, all these previous studies included combined valve procedures. This strongly limits any conclusion about the specific impact of the TV procedure on long-term outcome. In the present study on isolated TV procedures, we found no significant difference in survival between repair and replacement (data not shown). Limitations: We acknowledge some important limitations of our study. First, the study was a retrospective, non-randomized study comparing two groups with relatively small numbers of patients. The decision on surgical procedure (beating or non-
406
BARAKI, ET AL
beating, repair or replacement, and the method of repair) was based on surgeon preference and experience, which resulted in a selection bias and explains the difference in the preoperative data between the 2 groups as shown in Table I. Propensity score matched analysis would have made the results of this study more clear. Unfortunately, we failed to match the patients by propensity score because of the small number of patients and relatively strong bias in patient selection. Moreover, although we found a significant difference in the reoperation free survival rate, the number of patients was too small to detect the risk factors for reoperation. Secondly, we did not analyze more detailed early postoperative data such as catecholamine dosage, peak creatine kinase isozyme MB values, and peak troponin T values to further evaluate the subclinical potential benefits of the beating heart technique. This was due to inconsistent laboratory data acquisition during the observation period. Conclusion: In this study of 92 consecutive patients who underwent isolated TV surgery, no beneficial effect of beating heart surgery was found. Based on the lack of evidence for benefits of beating heart surgery, we recommend using cardioplegic arrest for isolated tricuspid valve surgery, especially in patients without left ventricular hypertrophy or poor heart function, considering the relatively short cross clamp times in the majority of these patients. We also found that the EuroSCORE and the preoperative heart rhythm are significant predictors in terms of long-term survival after isolated TV surgery. Atrial fibrillation and pacemaker dependency are associated with poor survival. Surgical intervention should be considered early in the course of the disease before irreversible heart failure or secondary end-organ problems occur.
Acknowledgments We gratefully acknowledge the assistance of Mrs. Pia Kokemohr for data acquisition and follow-up of patients.
Disclosure Conflict of interest: Baraki and other co-authors have no con-
flict of interest.
References
1. McGrath LB, Gonzalez-Lavin L, Bailey BM, Grunkemeier GL, Fernandez J, Laub GW. Tricuspid valve operations in 530 patients. Twenty-five-year assessment of early and late phase events. J Thorac Cardiovasc Surg 1990; 99: 124-33. 2. Dalrymple-Hay MJ, Leung Y, Ohri SK, et al. Tricuspid valve replacement: bioprostheses are preferable. J Heart Valve Dis 1999; 8: 644-8. 3. McCarthy PM, Bhudia SK, Rajeswaran J, et al. Tricuspid valve repair: durability and risk factors for failure. J Thorac Cardiovasc Surg 2004; 127: 674-85. 4. Murashita T, Okada Y, Kanemitsu H, et al. Fate of functional tricuspid regurgitation after mitral valve repair for degenerative metral regurgitation. Circ J 2013; 77: 2288-94. 5. Izumi C, Miyake M, Takahashi S, et al. Progression of isolated tricuspid regurgitation late after left-sided valve surgery. Circ J 2011; 75: 2902-7.
Int Heart J July 2015
6. Rogers JH, Bolling SF. The tricuspid valve: current perspective and evolving management of tricuspid regurgitation. Circulation 2009; 119: 2718-25. (Review) 7. Capoun R, Thomas M, Caputo M, Asimakopoulos G. Surgical treatment of tricuspid valve endocarditis: a single-centre experience. Perfusion 2010; 25: 169-73. 8. Baraki H, Saito S, Al Ahmad A, et al. Surgical treatment for isolated tricuspid valve endocarditis- long-term follow-up at a single institution. Circ J 2013; 77: 2032-7. 9. Cohn LH. Tricuspid regurgitation secondary to mitral valve disease: when and how to repair. J Card Surg 1994; 9: 237-41. 10. Guenther T, Noebauer C, Mazzitelli D, Busch R, Tassani-Prell P, Lange R. Tricuspid valve surgery: a thirty-year assessment of early and late outcome. Eur J Cardiothorac Surg 2008; 34: 402-9. 11. Singh SK, Tang GH, Maganti MD, et al. Midterm outcomes of tricuspid valve repair versus replacement for organic tricuspid disease. Ann Thorac Surg 2006; 82: 1735-41. 12. Ratnatunga CP, Edwards MB, Dore CJ, Taylor KM. Tricuspid valve replacement: UK Heart Valve Registry mid-term results comparing mechanical and biological prostheses. Ann Thorac Surg 1998; 66: 1940-7. 13. Irwin RB, Luckie M, Khattar RS. Tricuspid regurgitation: contemporary management of a neglected valvular lesion. Postgrad Med J 2010; 86: 648-55. (Review) 14. Sha JM, Yan ZY, Zhu ZY, et al. Early and midterm results of repair of Ebstein’s anomaly with autologous pericardium. Thorac Cardiovasc Surg 2011; 59: 287-92. 15. Filsoufi F, Salzberg SP, Abascal V, Adams DH. Surgical management of functional tricuspid regurgitation with a new remodeling annuloplasty ring. Mt Sinai J Med 2006; 73: 874-9. 16. Matsumoto Y, Watanabe G, Endo M, Sasaki H, Kasashima F, Kosugi I. Efficacy and safety of on-pump beating heart surgery for valvular disease. Ann Thorac Surg 2002; 74: 678-83. 17. Salerno TA, Panos AL, Tian G, Deslauriers R, Calcaterra D, Ricci M. Surgery for cardiac valves and aortic root without cardioplegic arrest (“beating heart”): experience with a new method of myocardial perfusion. J Card Surg 2007; 22: 459-64. 18. Salerno TA, Suarez MR, Panos AL, et al. Efficacy, feasibility, and pitfalls of transseptal approach in beating-heart mitral valve surgery. J Card Surg 2009; 24: 495-8. 19. Ricci M, Macedo FI, Suarez MR, Brown M, Alba J, Salerno TA. Multiple valve surgery with beating heart technique. Ann Thorac Surg 2009; 87: 527-31. 20. Mo A, Lin H, Wen Z, Lu W, Long X, Zhou Y. Efficacy and safety of on-pump beating heart surgery. Ann Thorac Surg 2008; 86: 1914-8. 21. Mo A, Lin H. On-pump beating heart surgery. Heart Lung Circ 2011; 20: 295-304. (Review) 22. Macedo FI, Rodriguez Y, Salerno TA. Myocardial preservation: beating heart techniques. Semin Thorac Cardiovasc Surg 2011; 23: 314-7. 23. Pfannmüller B, Davierwala P, Misfeld M, Borger MA, Garbade J, Mohr FW. Postoperative outcome of isolated tricuspid valve operation using arrested-heart or beating-heart technique. Ann Thorac Surg 2012; 94: 1218-22. 24. Miyairi T, Miura S, Taketani T, et al. Intraoperative assessment for mitral valve competency in a beating heart under retrograde coronary perfusion. Int Heart J 2013; 54: 192-5. 25. Ji Q, Mei Y, Wang X, Ding W. On-pump versus off-pump coronary artery bypass surgery in high-risk patients. Int Heart J 2014; 55: 484-8. 26. Romano MA, Haft JW, Pagani FD, Bolling SF. Beating heart surgery via right thoracotomy for reoperative mitral valve surgery: a safe and effective operative alternative. J Thorac Cardiovasc Surg 2012; 144: 334-9. 27. Messika-Zeitoun D, Thomson H, Bellamy M, et al. Medical and surgical outcome of tricuspid regurgitation caused by flail leaflets. J Thorac Cardiovasc Surg 2004; 128: 296-302. 28. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham study.
Vol 56 No 4
BEATING VS. ARRESTED HEART TV SURGERY
N Engl J Med 1982; 306: 1018-22. 29. Ngaage DL, Schaff HV, Barnes SA, et al. Prognostic implications of preoperative atrial fibrillation in patients undergoing aortic valve replacement: is there an argument for concomitant arrhythmia surgery? Ann Thorac Surg 2006; 82: 1392-9. 30. Wang B, Xu ZY, Han L, Zahng GX, Lu FL, Song ZG. Impact of preoperative atrial fibrillation on mortarity and cardiovascular outcomes of mechanical mitral valve replacement for rheumatic mitral valve disease. Eur J Cardiothorac Surg 2013; 43: 513-9. 31. Saxena A, Dinh D, Dimitriou J, et al. Preoperative atrial fibrillation is an independent risk factor for mid-term mortality after concomitant aortic valve replacement and coronary artery bypass graft surgery. Interact Cardiovasc Thorac Surg 2013; 16: 488-94. 32. Rankin JS, Hammill BG, Ferguson TB Jr, et al. Determinants of operative mortality in valvular heart surgery. J Thorac Cardiovasc Surg 2006; 131: 547-57. 33. Bonow RO, Carabello BA, Chatterjee K, et al; 2006 Writing Com-
407
mittee Members; American College of Cardiology/American Heart Association Task Force. 2008 Focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation 2008; 118: e523-661. 34. Pfannmüller B, Moz M, Misfeld M, et al. Isolated tricuspid valve surgery in patients with previous cardiac surgery. J Thorac Cardiovasc Surg 2012; 146: 841-7. 35. Moraca RJ, Moon MR, Lawton JS, et al. Outcomes of tricuspid valve repair and replacement: a propensity analysis. Ann Thorac Surg 2009; 87: 83-8.
