|Year : 2016 | Volume
| Issue : 3 | Page : 668-673
The short-term outcome of redo-mitral valve surgery: emergency versus elective
Ahmed L Dokhan1, Ali H Taher2, Islam M Ibrahim1, Ayman M Asfour2
1 Cardiothoracic Surgery Department, Faculty of Medicine, Menoufia University, Menoufia, Egypt
2 Cardiothoracic Surgery Department, National Heart Institute, Cairo, Egypt
|Date of Submission||11-Jul-2015|
|Date of Acceptance||04-Sep-2015|
|Date of Web Publication||23-Jan-2017|
Ayman M Asfour
Cardiothoracic Surgery, Department, National Heart Institute, Cairo, 11599
Source of Support: None, Conflict of Interest: None
The aim of this study was to investigate the overall outcome of adult patients undergoing redo-mitral valve replacement as emergency cases against elective cases.
Patients with mechanical prosthetic heart valves are at risk of reoperation. The reported risk of mortality may be as low as 5.4-11% for elective reoperation, but as high as 38-61.5% for emergency procedures.
Patients and methods
Forty patients who had undergone previous mitral valve replacement were admitted for redo-mitral valve replacement during the period between May 2011 and May 2013 at the National Heart Institute. They were divided into two groups: group A: 20 patients were admitted from the ER as emergency cases; group B: 20 patients were admitted from outpatient clinics as elective cases. A mechanical valve was inserted with horizontal mattress pledgeted nonabsorbable sutures. Sutures were placed from left atrium to left ventricle. Tricuspid valve incompetence, if present, was corrected by tricuspid valve repair (De-Vega suture).
The hospital mortality was 20%. There was no effect regarding age, sex, cardiac rhythm, number of previous operations, type of the previous prosthesis, and interval from last implantation. Taking into consideration that mortality was higher in the emergency group, the New York Heart Association (NYHA) functional class, left ventricular end systolic diameter, left ventricular end diastolic diameter, redo-cardiac surgery sternotomy, and adhesiolysis carry a significant risk of catastrophic bleeding, especially with the rush accompanying hemodynamic instability. Infective cardiac tamponade, permanent pacemaker, residual infective endocarditis, the need for dialysis, and cerebrovascular accidents were not statistically significant.
When significant valve dysfunction is first noted, reoperation should be undertaken to minimize the operative risk to avoid mortality and postoperative morbidities.
Keywords: elective, emergency, mitral, outcome, redo operation
|How to cite this article:|
Dokhan AL, Taher AH, Ibrahim IM, Asfour AM. The short-term outcome of redo-mitral valve surgery: emergency versus elective. Menoufia Med J 2016;29:668-73
|How to cite this URL:|
Dokhan AL, Taher AH, Ibrahim IM, Asfour AM. The short-term outcome of redo-mitral valve surgery: emergency versus elective. Menoufia Med J [serial online] 2016 [cited 2020 Apr 4];29:668-73. Available from: http://www.mmj.eg.net/text.asp?2016/29/3/668/198752
| Introduction|| |
Patients undergoing valve reoperations have a diverse and complex clinical profile. Thrombi may form on mechanical valves and cause orifice obstruction, leaflet malcoaptation, and acute valvular dysfunction. Hence, redo-valve surgery is associated with a higher operative mortality rate than first-time valve surgery, but certain risk factors may be preventable. With advances in medical therapy and life expectancy, reoperation to replace dysfunctional mechanical heart valve prosthesis is an increasingly common procedure and there have been gradual decreases in the perioperative risk for redo-valve surgery over the past two decades, likely due to increased surgical experience, better myocardial protection, and improved patient management. However, mortality rates remain higher than with first-time valve-replacement surgery  . Mechanical valves have excellent durability, but also have a considerable risk of primary valve failure, prosthetic valve endocarditis, prosthetic valve thrombosis, thromboembolism, and mechanical hemolytic anemia. In addition, because many of these patients require long-term anticoagulation, anticoagulant-related hemorrhage may occur.
In elective cases of redo-mitral valve replacement (nonobstructive thrombus), treatment consists of a short course of intravenous heparin with close echocardiographic follow-up in addition to readjustment of the warfarin dose and addition of aspirin  . However, if the medical treatment is unsuccessful, surgery should be considered. Surgery should be initially decided in patients with large (>5 to 10 mm as determined by transesophageal echocardiographic study) or mobile thrombi. Thrombolysis with, for example, streptokinase or urokinase, can also be used according to some studies in these cases.
In emergency cases of redo-mitral valve replacement (obstructive thrombus), surgery is the treatment of choice in patients with thrombosis causing valve obstruction. 'In a recent series, operative mortality was 4-5% for patients with New York Heart Association (NYHA) class III or lower, whereas it reached 15-20% in patients with class IV. The intervention may involve simple thrombectomy or valve replacement  .
| Patients and methods|| |
Patients were divided into two groups:
(1) Group A: 20 patients were admitted from the ER as emergency cases
(2) Group B: 20 patients were admitted from outpatient clinics as elective cases.
Patients who underwent mitral-valve replacement redo surgery in cases of surgery for prosthetic endocarditis, paraprosthetic leak, structural valve degeneration, or prosthetic valve thrombosis were included.
Patients with the following conditions were excluded: impaired renal or liver functions, previous cerebrovascular disease, impaired respiratory functions (e.g., chronic obstructive pulmonary disease, emphysema, suppurative lung disease, etc.), concomitant coronary artery disease, or aortic valve disease.
On presentation, all patients were evaluated by a detailed medical history, the operative data including the valve type and size, the type of incision (sternotomy vs. thoracotomy), and the international normalized ratio (INR). It should also be noted that symptoms in these patients have a very wide range, from mildly symptomatic patients to patients presenting with cardiogenic shock with very poor condition. However, some patients came as a result of echocardiographic evidence of valve dysfunction, while they are completely asymptomatic.
Detailed clinical examination was performed, including checking of the access of the previous operation (sternotomy vs. thoracotomy) and cardiac auscultation including valve sounds, any abnormal sounds, and murmurs. It is believed that tilting disc and bileaflet valves have a loud, high-frequency, metallic closing sound. Checking vital signs plays a role in the suspicion of valve dysfunction. Heart rate and its regularity, assessed as tachycardia in a previously controlled patient's rate, may signify a new event: heart failure or fever in prosthetic valve endocarditis. Hypotension may occur in cases of cardiogenic shock.
Routine laboratory investigations performed included a complete blood count, urea and creatinine, and a liver enzymes and bleeding profile. Chest radiograph (posteroanterior and lateral views) was obtained, if the condition allowed, to assess the chest, the number of stainless steel wires of previous sternotomy, and the relation between the heart and the sternum. ECG was performed to detect the rhythm of the patient's heart rate. Echocardiography was performed to assess leaflet mobility, the presence of thrombi, vegetations, paravalvular leak, valve dehiscence, pulmonary artery pressure, chamber dimensions, and the ejection fraction. Fluoroscopy was performed if the echocardiography was not conclusive for valve mobility in cases with suspicion of diminished leaflet excursion or high transprosthetic gradients. Blood culture was also performed in suspected cases of endocarditis. A full echocardiographic evaluation includes two-dimensional imaging of the prosthetic valve, assessment of leaflet mobility, measurement of transprosthetic gradients across the valve and the effective orifice area, estimation of the degree of paravalvular leak, evaluation of LV dimensions and systolic function, and calculation of systolic pulmonary arterial pressure.
Under general anesthesia in the supine position, all operations were carried out through median sternotomy after exposure of the femoral artery and vein to be used if needed. The heart was mobilized by gentle dissection over the right atrium, leaving the left side without dissection, unless needed to avoid bleeding from exposed raw surface. Access to the valve was obtained through the left atriotomy or trans-septal approach in cases with a small left atrium. The valve was removed gently, together with any thrombus or pannus, and the native ring was debrided from any previous stitches or fibrous tissue or vegetations (in case of infective endocarditis).
Patients were transferred to the ICU on inotropic support as needed.
Postoperatively, all patients received permanent anticoagulation with warfarin if there was no evidence of active bleeding. The INR was maintained between 2.5 and 3.5 in all patients. The fluctuant levels of INR required more frequent follow-up and were arranged by either telephone interviews or in the hospital.
Echo was scheduled for all patients 2 weeks and 6 months postoperatively.
Patients' categorical predictor variables and outcomes were analyzed using the Pearson χ-test for independence. Statistical analysis was performed using Microsoft Office Excel 2010 and SPSS (version 20, 2011, SPSS Inc., Chicago, Illinois, USA). P value less than 0.05 was considered statistically significant.
| Results|| |
In the emergency group, with regard to hemodynamics, five patients were unstable (12.5%) (five patients arrived in the operation room under shock, with pale cold skin, low urine output, tachypnea, tachycardia, and a reduced conscious level) and five patients were on inotropes (12.5%). A total of 30 patients were stable (75%): 20 cases in the elective group and 10 in the emergency group ([Table 1]). With regard to the NYHA classification, two cases (5%) from the elective group were NYHA III, whereas it was six cases (15%) in the emergency group [five cases (12.5%) NYHA IV and one case (2.5%) NHYA III]. Regarding the number of previous mitral valve replacements, in the elective group, 18 patients had undergone previous mitral valve replacement once and two patients twice, whereas in the emergency group, 15 patients had undergone previous mitral valve replacement once before, four patients twice, and one patient thrice ([Table 2]).
|Table 1 Hemodynamic stability condition preoperatively in both groups (elective and emergency) |
Click here to view
|Table 2 The number of redo cases in both groups (elective and emergency) |
Click here to view
Indications for reoperation ([Table 3]) were as follows: in the emergency group, 12 patients had thrombosis, four had paravalvular leak, four had endocarditis, and no patients had a degenerative valve; in the elective group, 14 patient had thrombosis, five had a paravalvular leak, and one patient had a degenerative valve. Tricuspid regurge was found in all preoperatively assessed cases. The cross-clamp time was a little shorter in elective cases (55-145 min) compared with emergency cases (65-185 min), and the bypass time was 75-240 min for elective cases and 90-260 min for emergency cases.
|Table 3 Indication of reoperation in both groups (elective and emergency) |
Click here to view
Catastrophic blood loss occurred in six patients (15%), all in the emergency group. Intraoperative mortality occurred in one case (2.25%) in the elective group and in four cases (10%) in the emergency group. Hospital mortality occurred in one case (2.25%) in the elective group and in two cases in the emergency group due to poor contractility. A second look was needed for one case in elective group due to bleeding, and one case in the emergency group (open chest).
Immediate postoperative data
Prolonged ICU stay was needed in two cases in the elective group (one due to chest infection, the other due to weaning of inotropes), and in three cases in the emergency group (two due to chest infection and one due to weaning of inotropes).
With regard to postoperative kidney functions, one case in the emergency group needed dialysis.
A permanent pacemaker was needed for one case in the emergency group. Residual infective endocarditis remained in one case (fungal type) in the emergency group. One case had cardiac tamponade (the patient was discharged and returned 3 weeks later to the emergency department with severe dyspnea; Echo revealed a massive effusion that was drained with subxiphoid incision), also in the emergency group.
The ejection fraction and the left-ventricle dimensions improved in most of the cases. Residual tricuspid regurge occurred in six patients (15%) in each group, making up a total of 12 patients (30%).
Gradients also decreased in all patients: the maximum gradient ranged from 6 to 12 mmHg in the elective group and was almost the same in the emergency group (6-14 mmHg). Also, the mitral-valve area ranged from 1.8-3.1 cm 2 in both groups ([Table 4]). Mortality due to low left-ventricular ejection fraction (<35%) and increased LVEDD more than 50 mm was noticed in one case in the elective group and two cases in the emergency group.
|Table 4 Left ventricle dimensions and mitral valve gradients postoperatively |
Click here to view
Overall, elective cases had a much better outcome considering mortality (two cases in the elective group compared with six cases in the emergency group) and morbidities.
| Discussion|| |
In this study, the overall hospital mortality was eight cases (20%): two cases in the elective group (5%) (one intraoperative mortality and one ICU mortality) compared with six cases in the emergency group (15%) (four cases of intraoperative mortality and two cases of ICU mortality).
De Almeida Brandγo et al.  reported a hospital mortality of 10.9% for emergency redo cases. In another study conducted in 2002, Kumar and associates reported a mortality of 11% in emergency redo-mitral valve operation. Overall, the operative mortality was 8.4% in elective redo-mitral operations. Wauthy et al.  observed a mortality rate of 8% in elective redo-valve surgery. Such differences in mortality may be, as mentioned by Wauthy, related to technological evolutions of cardiac critical care units (e.g. defibrillation patches, improved extracorporeal circulatory support technology, the use of extra corporeal membrane oxygenation, the cell saver, etc.) and increased surgical experience.
In this study, mortality in relation to the NYHA classification were two cases (5%) with NYHA III in the elective group and six cases (15%) in the emergency group [five cases (12.5%) with NYHA IV and one case (2.5%) with NHYA III]. NYHA functional class IV was also a risk factor in short-term survival as mentioned by Akay et al.  and others in many studies. We concluded that NYHA classification is a statistically significant factor for mortality in both groups. Vohra et al.  also confirmed that the NYHA functional class was highly significant for operative mortality; as operative mortality in their study was 4% for functional classes I through III and 19% for functional class IV, they reported the NYHA functional class as an important risk factor for hospital mortality. NYHA FC is considered as the most frequently quoted risk factor associated with death in redo-valve surgery. This was due to the fact that mortality in their study reached up to 30% with stage IV (half cases were emergency) compared with less than 10% in stages II and III (all cases were elective).
Considering morbidities, reopening was performed in one case (2.5%) in the elective group and in one case (2.5%) that was left open chest in the emergency group. This morbidity was not statistically significant between the two groups.
Akay and colleagues reported bleeding in 5.6% of their patients with equal incidence in both emergency and elective cases. Potter et al.  reported excessive postoperative bleeding (more than 1000 ml in the first 24 postoperative hours) in 14.5% (9% were for emergency group) of the patients, while re-exploration was performed in only 8% of them.
In this study, there was a need for dialysis in one case (2.5%) in the emergency group due to constant increase in creatinine and potassium levels. Akay and colleagues reported 14.2% with postoperative renal dysfunction 10% for emergency group. Preoperative renal impairment, CVS, prolonged bypass time, and cross-clamp time are risk factors for postoperative renal dysfunction.
In this study, a permanent pacemaker was installed in one case (2.5%) in the emergency group due to complete heart block that did not recover after 14 days. This morbidity was not statistically significant between the two groups.
Potter and colleagues reported complete heart block and bradyarrhythmia in 3% of the cases of redo-mitral valve surgery: only 0.3% of the cases needed a permanent pacemaker (one case after redo aorta and mitral).
In our study, the cause of this prolonged time for bypass was in most cases due to the extra time needed for circulatory support due to associated left ventricular dysfunction. The mean time for the cross-clamp in elective cases was 100 mins and for emergency cases 125 mins, Meanwhile the bypass times were 157.5 and 175 mins, for elective and emergency cases respectively. This was not statistically significant for both groups. Potter et al. stated that prolonged bypass and cross-clamp times are predictors of mortality. De Almeida Brandγo and colleagues mentioned that a cross-clamp time longer than 120 min was among the intraoperative variables associated with a higher hospital mortality. Global myocardial ischemic time alone was a strong predictor of hospital mortality in many studies. In contrast, others denied both factors as being significant predictors of hospital mortality.
In our study, thrombosis occurred in 26 patients (65%): 14 (35%) in the elective group and 12 (30%) in the emergency group; paravalvular leak occurred in nine patients (22.5%): five (12.5%) in the elective group and four (10%) in the emergency group; endocarditis occurred in four patients (10%), and all were in the emergency group; there was one case of degenerative valve in the elective group (2.5%). Indication of reoperation had no statistical significance regarding the outcome for both groups. De Almeida Brandγo and Vohra mentioned that the indication for surgery had no impact on in-hospital mortality. According to Maciejewski et al.  , operative mortality was significantly higher in those patients who underwent reoperation because of prosthetic endocarditis. The presence of valve infection is also a predictor. Other observations suggest that the factors responsible for higher mortality are active infective endocarditis and valve thrombosis.
In our study, 33 patient (82.5%) had undergone the operation once before [18 cases (45%) in the elective group and 15 cases (37.5%) in the emergency group], and six patients (15%) had undergone the operation twice before [two cases (5%) in the elective group and four cases (10%) in the emergency group]. One case (2.5%) in the elective group had undergone the operation three times before. The number of prior operations did not show a significant effect on the outcome on both groups.
Although there was a trend to higher operative mortality with increasing number of prior cardiac operations, this factor was not significant in the multivariable analysis. Some reported that second reoperations were not more risky than first reoperations, but by the third reoperation, risks were high for all subgroups. This was confirmed by Beghi et al.  , where the number of previous reoperations was an independent determinant for reoperation.
Our study suggests that left ventricular dysfunction was associated with higher hospital mortality in both groups [one case (2.5%) in the elective group and two cases (5%) in the emergency group]. This was statistically significant. A low left-ventricular ejection fraction (<35%) and increased LVEDD more than 50 mm have been reported to be significant predictors of mortality by Akay and colleagues. According to Maciejewski and colleagues, operative mortality was significantly higher in those patients who had impaired left ventricular function.
| Conclusion|| |
The NYHA functional class, increased LVEDD and LVESD, and redo-cardiac surgery sternotomy (where adhesiolysis carries a significant risk of catastrophic bleeding, especially with the rush accompanying hemodynamic instability), infective endocarditis, left ventricular dysfunction, and ejection fraction less than 35 were the main parameters affecting the outcome of surgery, with higher mortality and morbidities in cases involving these criteria in the emergency group.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Tang GHL, Maganti M, David TE, Feindel CM, Scully HE, Borger MA. Effect of prior valve type on mortality in reoperative valve surgery. Ann Thorac Surg 2007; 83
Pibarot P, Dumesnil JG. Prosthesis-patient mismatch: definition, clinical impact, and prevention. Heart 2006; 92
Vesey JM, Otto CM. Complications of prosthetic heart valves. Curr Cardiol Rep 2004; 6
De Almeida Brandão CM, Pomerantzeff PMA, Souza LR, Tarasoutchi F, Grimberg M, Ramires JAF, Almeida de Oliveira S. Multivariate analysis of risk factors for hospital mortality in valvular reoperations for prosthetic valve dysfunction. Eur J Cardiothorac Surg 2002; 22
Wauthy P, Goldstein JP, Demanet H, Deuvaert F. Redo valve surgery nowadays: what have we learned?. Acta Chirurgica Belgica 2003; 103
Akay TH, Gultekin B, Ozkan S, Aslim E, Uguz E, Sezgin A, Aslamaci S. Mitral valve replacements in redo patients with previous mitral valve procedures: mid-term results and risk factors for survival. J Card Surg 2008; 23
Vohra HA, Whistance RN, Roubelakis A, Burton A, Barlow CW, Tsang GMK, Ohri SK. Outcome after redo-mitral valve replacement in adult patients: a 10-year single-centre experience. Interact Cardiovasc Thorac Surg 2012; 14
Potter DD, Sundt TM, Zehr KJ, Dearani JA, Daly RC, Mullany CJ, Orszulak TA. Risk of repeat mitral valve replacement for failed mitral valve prostheses. Ann Thorac Surg 2004; 78
:67-72. discussion 67-72.
Maciejewski M, Piestrzeniewicz K, Bielecka-D¹browa A, Piechowiak M, Jaszewski R. Redo surgery risk in patients with cardiac prosthetic valve dysfunction. Arch Med Sci 2011; 7
Beghi C, De Cicco G, Nicolini F, Ballore L, Reverberi C, Gherli T. Cardiac valve reoperations: analysis of operative risk factors in 154 patients. J Heart Valve Dis 2002; 11
[Table 1], [Table 2], [Table 3], [Table 4]