Hassan Houmaida

Acute aortic valve thrombosis secondary to recombinant factor VIIa.

A60-year-old male patient was admitted on an elective basis for coronary artery bypass grafting. His past medical history included hypertension and multiple sclerosis. He had experienced an acute inferior myocardial infarction 1 month earlier. Coronary angiography revealed severe triple-vessel disease. Transthoracic echocardiography disclosed a normal aortic valve and an ejection fraction of 45%. The patient had normal hepatic and renal function and a normal coagulation profile before the operation. Cardiopulmonary bypass was established through a median sternotomy. A saphenous vein was grafted onto the right coronary artery, and the left internal thoracic artery was anastomosed to the left anterior descending artery. The obtuse marginal artery was not identified during the operation and was therefore left ungrafted. Attempts to wean the patient from cardiopulmonary bypass were unsuccessful, and femorofemoral extracorporeal membrane oxygenation (ECMO) was instituted. Electrocardiography revealed ST segment elevation in lateral chest derivations. The patient was

Bartonella as a Cause of Mechanical Prosthetic Aortic Root Endocarditis

Bartonella henselae infection is typically associated with cat scratch disease. This microorganism can also lead to culture-negative infective endocarditis in immunocompromised patients. We present a rare case of a previously healthy 65-year-old man with B. henselae-associated endocarditis of a prosthetic aortic root. All blood cultures, as well as cultures of the resected aortic valve vegetations, remained negative. Polymerase chain reaction with specific bacterial primers with DNA sequencing was used to identify B. henselae as the etiologic agent. This was successfully managed by an aortic root re-replacement using a mechanical conduit, reimplantation of coronaries ostia, and antibiotic therapy.

eComment. Are mechanical valves better than bioprostheses in patients on dialysis

We read with great interest the article by Pai et al. regarding the best valve substitute in patients on dialysis [1]. They included in the results of their research, seven retrospective studies and one meta-analysis. However, we found 4 other relevant articles investigating the same problem. In order to be exhaustive, we will summarize the relevant results of these studies and highlight the safety of implanting bioprosthesis in patients on chronic dialysis. The standard of valve selection has changed over time. It has long been believed that tissue valves undergo premature degeneration due to the derangements in calcium metabolism in patients with end-stage renal disease. This is based on largely anecdotal case reports using first generation bioprostheses. In 1998, ACC/AHA guidelines recommended the use of mechanical valves in patients on dialysis. Accumulating data supporting the very low incidence of rapid tissue valve degeneration in dialysis patients had been taken into consideration, and the latest ACC/AHA practice guidelines do not specify the best choice for valve replacement in dialysis patients. Only four cases of structural valve deterioration (SVD) requiring reoperation were identified from the meta-analysis [1], ranging from 10 to 96 months after the initial valve replacement surgery. However, conclusions on the long-term performance of tissue valves in this patient population cannot be drawn. Bleeding was the most common valve-related complication, and represented a major drawback of mechanical valves. Lucke et al. [2] reviewed 19 consecutive patients with end-stage renal disease from a single institution who had undergone aortic, mitral or aorto-mitral valve replacement, 9 had a bioprosthetic valve and 10 a mechanical valve. Mechanical valve patients had a significantly higher rate of postoperative cerebrovascular events or bleeding complications. No subsequent reoperations were required for biological valve failure. The overall estimated Kaplan-Meier survival was 42% ± 14% at 60 months. Kaplon et al. [3] from The Cleveland Clinic Foundation, found comparable results for both types of valves when reviewing 42 patients on preoperative dialysis undergoing valve replacements. Seventeen patients received mechanical valves and 25 received bioprosthesis. Four patients with a bioprosthesis required reoperation, one of whom experienced mitral bioprosthesis degeneration. Prosthetic valve-related complications and survival were similar for both mechanical and bioprosthetic valves. Toole et al. [4] reviewed 50 dialysis patients undergoing left-sided valve replacement. The tissue valve group had significantly higher Kaplan-Meier freedom from valve-related morbidity and mortality at three years. Freedom from reoperation was not significantly different. Umezu et al. [5] analyzed data from 63 consecutive dialysis patients who underwent valvular surgery. The mechanical group had a higher rate of bleeding events. There was no case of SVD up to the 5-year follow-up. However, both mechanical and bioprosthetic valve patients had similar survival and event-free rates. It can be concluded that dialysis patients after cardiac valve replacement suffer poor mid- and long-term survival. Therefore, surgeons should not hesitate to implant bioprosthetic valves because SVD will be uncommon in this patient population. Prosthesis selection should be based on the same criteria used for non-dialysis patients. Conflict of Interest: None declared

eComment. Primary cardiac malignancy

An interesting case report is presented by Liu et al [1]. A 41-year old immunocompetent female patient underwent a surgical excision of a primary cardiac B-cell lymphoma (74 mm x 49 mm) from the right atrium. Only 25% of primary cardiac tumours are malignant, and sarcomas represent more than two-thirds of these malignancies. Primary cardiac lymphoma (PCL) is the second most common malignancy affecting the heart. Thus, PCL is a very rare entity and is classified as primary lymphomas inside the heart chamber, primary lymphomas developing in the epicardial space and primary large cell lymphomas involving the pericardium. PCL is defined by the absence of lymphoma outside the pericardium after a complete clinical and radiological examination and by the confinement of the bulk of the tumour to the pericardial sac. Cardiac metastases of extra-cardiac lymphomas are far more common than PCL. This case report has an interesting feature that could be of use to all of us when faced with a symptomatic right a trial tumour in a young patient. Cardiac surgeons should be highly suspicious of primary cardiac malignancy when the signs are suggestive. Right-sided cardiac tumours are usually malignant and appear as infiltrative, bulky and fast- growing masses that develop in an outward fashion. They do not present with congestive heart failure until the later stages of disease progression. The diagnostic algorithm of choice when suspecting a primary cardiac malignancy is to perform after transthoracic echocardiography, cardiac computed tomography and cardiac magnetic resonance imaging [2]. Cardiovascular magnetic resonance can recognize, with high specificity, non-surgical cardiac masses: pseudo-tumours, lipomas, thrombi, and lipomatous hypertrophy. Most other cardiac tumours will necessitate tissue sample diagnosis to help in the establishment of a treatment plan. Should the use of multiple imaging techniques fail to identify the cardiac mass at the outset, a biopsy (percutaneous endomyocardial biopsy or a biopsy by thoracotomy) is best obtained if it can be done safely [3]. This will not only diagnose the tumour but will exclude benign masses and other malignant tumours, such as lymphoma, that are best treated non-surgically. Treatment of PCL is usually bimodal [4] with aggressive systemic chemotherapy (cyclophosphamide, doxorubicin, vincristine, and prednisone) as well as radiation therapy. Conflict of Interest: None declared

eComment. Cardiac tamponade as a manifestation of acute mediastinitis

Valenzuela and colleagues [1] report a case of a 53-year old woman with an iatrogenic right coronary artery dissection, who underwent an urgent off-pump coronary artery bypass grafting to the right coronary system. One week later, she was operated on for a compressive purulent pericardial effusion without signs of wound infection. Unfortunately, she died because of therapy-refractory multiorgan failure. This report highlights the difficulty in making a timely diagnosis of this particular condition. Post-sternotomy mediastinitis is an ominous complication carrying a high risk of in-hospital mortality. An abundance of risk factors for the development of anterior mediastinitis has been identified and classified as preoperative (morbid obesity, osteoporosis, chronic obstructive pulmonary disease, diabetes, Staphylococcus-positive nasal swab, mediastinal radiotherapy and immunosuppressive intake), intraoperative (break in sterility, paramedian sternotomy, bone fracture with sternal retractor, technical mistakes in sternal closure, bilateral harvesting of thoracic arteries and prolonged operative time) and postoperative factors (prolonged time on ventilator and acute delirium syndrome) [2-4]. Despite regular use of prophylactic intravenous antibiotics and preventive measures, post-sternotomy mediastinitis occurs in a significant number of patients undergoing cardiac surgery. The Centers for Disease Control definition for surveillance of surgical site infections identifies three categories of surgical site infection [5]: superficial incision surgical site infection, deep incision surgical site infection and organ surgical site infection. According to the preceding definition, the in-hospital acquired infection of the retrosternal/mediastinal space of this patient without incisional site purulent drainage is not considered as a surgical site infection. The aggressiveness of the infection in this case resembles the course of an untreated descending necrotizing mediastinitis arising from odontogenic or cervico-facial infection. Another similarity is the dissemination of the infection from pleural space to the pericardial space through the opening of the right pleura. Infection in the setting of descending mediastinitis may spread across fascial planes, thus cross-contaminating into the anterior, middle, and posterior mediastinum and even pleural spaces. As outlined in this case, mediastinitis remains a rare but devastating complication of open heart surgery. While prompt diagnosis and emergent surgical intervention are crucial for better outcome, the presenting symptoms can be easily confused with more common problems such as cardiac tamponade. Therefore, a high index of suspicion for mediastinitis is crucial for survival even in the absence of surgical site infection. Conflict of interest: none declared.

Left thoracotomy for coronary artery bypass grafting in a patient with an ileo-coloplasty.

A year-old male patient with a history of right ileo-coloplasty for oesophageal carcinoma five years earlier presented with chest pain. Transthoracic echocardiogram showed moderate left ventricular dysfunction (EF = 45%). Cardiac catheterisation revealed high-grade coronary artery lesions affecting the left main trunk, left anterior descending (LAD) artery and the obtuse marginal artery (Fig. 1A, arrows showing haustra of the colon). A computed tomography scan was performed to choose the best surgical approach (Fig. 1B, arrow depicting the neooesophagus lying behind the left side of the sternum). The patient underwent off-pump CABG through a left anterolateral thoracotomy using the fifth intercostal space; he was positioned with the left side elevated to approximately 45◦. Under single lung ventilation we successfully dissected the colon from the posterior aspect of the sternum, and subsequently the left internal thoracic artery was harvested under direct vision from a lateral approach. The pericardium was opened antero-laterally, and pericardial stay sutures were placed on both edges allowing good exposure of the target vessels. The left internal thoracic artery was anastomosed to the LAD. Progressive mobilisation of the apex to explore the obtuse marginal artery was followed by a severe hemodynamic instability. The procedure was carried out under femo-femoral cardiopulmonay bypass to assist the failing heart. The proximal anastomose of the saphenous vein used to bypass the

eComment. Catheter-induced acute aortic syndrome

We read with great interest the report by Attia et al [1]. The authors presented the case of a 75-year old patient with an iatrogenic intramural collection of contrast agent into the ascending aortic wall during cardiac catheterization. The patient was managed conservatively, and the follow-up at three months was unremarkable. However, we believe that there are some issues that need to be addressed. Acute aortic dissection is the most feared complication of cardiac catheterization, it occurs in up to 0.02% of diagnostic catheterizations and 0.07% of percutaneous coronary interventions [2]. In the majority of cases, there is an intimal tear in the vicinity of the ostium of the coronary artery and subsequently a bidirectional dissection, an antegrade dissection of the coronary artery and retrograde dissection of the ascending aorta. Risk factors for catheter-induced aortic dissection include: left main coronary artery intubation, use of special catheters (Amplatz-shaped catheters), aggressive handling of rigid guide wires, catheterization during acute myocardial infarction, vigorous contrast injections and profound cannulation of the catheter into the coronary ostia [3]. The management and outcome of catheter-induced acute aortic dissection depend on the status of the distal flow in the coronary artery and extent of the propagation of the dissection. Stenting of the coronary ostium to seal the primary entry tear has been reported as the best treatment option, thus allowing the aortic dissection to heal spontaneously [4]. However, the indications for surgical repair of the ascending aorta in otherwise clinically stable patients without malperfusion, tamponade or significant aortic regurgitation have not been clearly defined, with some operators adopting a conservative approach, and others advocating an early surgical intervention. Although conservative treatment of uncomplicated retrograde dissection with blood pressure control, analgesics, and close observation has been recommended, especially in high-risk patients or those with previous cardiac surgery [5]. The progression of these dissections is unpredictable and can be life-threatening. We have only one question for Attia et al. regarding the definition of intramural haematoma in this setting. How could the authors explain the possible mechanism of the intramural collection caused by contrast extravasation without an intimal tear? We think that this image is the result of stasis of contrast agent within the false lumen, and it represents a localized aortic dissection with residual contrast within the aortic tunica media passing through an intimal tear without an exit tear. Conflict of interest: none declared

eComment. Valve replacement in carcinoid heart disease

We read with great interest the paper by Mabvuure et al. regarding the durability of bioprostheses in patients with carcinoid valve disease [1]. Carcinoid heart disease occurs in the context of metastatic serotonin-producing neuroendocrine tumours in the liver and classically presents as a progressive dysfunction of the tricuspid and/or pulmonary valve due to endocardial plaque deposition, thus reducing the mobility of the right-sided valve leaflets [2]. Surgical valve replacement is the only effective treatment when symptoms of right heart failure emerge in this patient population. Mabvuure and colleagues included the results of 17 papers in their research, and the pooled data represented 51 patients with bioprostheses in the tricuspid position [1]. Recently, an additional retrospective study was published [3] and the purpose of this study was to assess the early and late outcomes of patients with carcinoid heart disease after valve replacement. In the above-mentioned study, three patients received one or two bioprostheses. However the remaining 16 patients underwent implantation of mechanical valve prostheses. Early postoperative mortality was 10% and survival rates at 1 and 5 years were 71 and 43% respectively. At the last follow-up, all survivors were in NYHA class I and echocardiography showed improvement of right ventricle function in the majority of patients. Although we recognize the efforts of Mabvuure and colleagues to identify the durability of biological valve in patients with carcinoid heart disease undergoing valve replacement by reviewing the relevant literature, we would like to point out an important point that they failed to include in their report. Classically, severe tricuspid regurgitation coexists with pulmonary valve stenosis in patients with symptomatic carcinoid valve disease. The simple surgical management consists in a tricuspid replacement in combination with pulmonary valvectomy. This will leave the patient with some degree of pulmonary incompetence while implanting one prosthetic valve. For patients with combined pulmonary and tricuspid stenosis, percutaneous balloon valvuloplasty is a feasible and effective alternative in cases with severe comorbidity or frailty [4]. In conclusion, surgeons should not hesitate to implant bioprostheses in this patient population. Accumulating data supporting the very low incidence of premature tissue valve degeneration in patients with carcinoid heart disease should be taken into consideration. Nevertheless, in patients where the carcinoid tumor is well-controlled and long-lasting survival is anticipated, a mechanical prosthesis may be implanted to avoid the possibility of carcinoid plaquing of the biological valve. However, in the setting of serious liver metastases, the use of a bioprosthesis could still be a better alternative. Conflict of interest: none declared.

eComment: do gentamicin-collagen sponges decrease the risk of mediastinitis?

We read with great interest the paper by Godbole and colleagues about the use of gentamicin-collagen sponges in the closure of sternal wounds in cardiac surgery to reduce wound infections [1]. In the results of their research, they have included preliminary conclusions from a controlled, prospective, randomized, double-blind, single-centre study [2]. Recently, the results of this investigation were fully published [3]. In order to be exhaustive, we have summarized the relevant results of the above-mentioned study. In this monocentric study, which was conducted in Germany, the gentamycin-collagen sponge group (n = 353) showed a significantly (p = 0.014) reduced risk of deep sternal wound infection as compared to the control group (n = 367) – that is, 0.56% and 3.52% at 30 days, respectively. The two types of sponges were implanted retrosternally without premoistening and could not be differentiated by the surgical team. There were no statistically significant differences between the two groups in terms of secondary end points such as superficial sternal wound infection, postoperative bleeding or transfusion of red cell units. The design of this surgical study has met the criteria of the Oxford Centre for Evidence-based Medicine for evidence grade 1b, which is the highest quality rating for an individual study in this field. Therefore, in discussing the clinical implications of their findings, the authors have suggested that routine prophylactic retrosternal placement of gentamicin-releasing sponge reduced the incidence of deep sternal wound infection. Anterior mediastinitis is reported to occur in 0.4 to 4% of median sternotomies [4] and this particularly grave infection is still responsible for significant mortality and morbidity. Many factors responsible for poststernotomy mediastinitis have been suggested and these have been classified as preoperative, intraoperative and postoperative factors. Every effort to reduce the incidence of poststernotomy mediastinitis is valuable. The implementation of this simple measure could bring particular benefits to high-risk patients. Future studies with larger numbers of patients and a longer follow-up will be required to validate the efficacy of the implantation of gentamycin-collagen sponge retrosternally and to develop evidence-based recommendations of highest statistical certainty. Conflict of Interest: None declared

eComment. Cardiopulmonary bypass and pregnancy

We read with great interest the article by Sepehripour et al. regarding the safety of performing cardiac surgery on cardiopulmonary bypass (CPB) during pregnancy [1]. They included in the results of their research three retrospective literature reviews (published in 1996 and 1998), three retrospective observational studies and eight case reports describing 10 parturients undergoing CPB. However, we found three additional relevant articles investigating the outcome of pregnancy after CBP. In order to be exhaustive, we will summarize the relevant results of the omitted papers. Thalmann et al. [2] performed a population-based study covering a period of 10 years in Austria and found 2 pregnant patients (at 36 and 32 weeks of pregnancy) with acute aortic dissection. Emergent surgery was successful for both patients and their offspring. The relevant result of the study is that pregnancy may not be a risk factor for acute aortic dissection and hence contradicts the findings of previous studies. Guo et al. [3] reported two cases of acute aortic dissection during the third trimester. The first was a 33-year old parturient with Marfan syndrome who underwent a successful Bentall procedure at 28 weeks of gestation, and the second was a 30-year old pregnant patient at 32 weeks of pregnancy who underwent an uneventful aortic root replacement. Both patients underwent a caesarean section immediately before CBP. We recently reported a successful surgical pulmonary embolectomy under CBP [4]. The patient was a 39-year old at 25 weeks of pregnancy who underwent an emergent surgery due to a massive pulmonary embolism with haemodynamic compromise. In this case of heparin-induced thrombocytopenia, we chose to carry out a surgical embolectomy because we were familiar with performing this procedure when fibrinolytic therapy is contraindicated. On the basis of accumulating evidence of encouraging results, we believe it is reasonably safe to perform cardiac surgery during pregnancy at tertiary centres with expertise using additional strategies to minimize maternal and foetal risks (normothermic CBP with high flow rate and left lateral recumbent position) [5]. The maternal mortality rate is reported to be comparable to that of CPB in non-pregnant women, except in the emergent cases. In the other hand, the foetal mortality rate decreases when cardiac surgery is postponed and the foetus is allowed to mature. Conflict of interest: none declared