Article ID: 2022-0009-CR
Patients with tracheostomy who undergo a full sternotomy have an increased risk of mediastinitis and sternal infection. This report describes a reoperative aortic valve replacement (re-AVR) for structural valve deterioration (SVD) through a lower hemisternotomy. This procedure was performed on a 71-year-old man with a tracheostomy who had previously undergone a Bentall procedure with a bioprosthetic valve to replace an enlarged ascending aortic aneurysm. Comorbidities included chronic renal failure requiring hemodialysis. Fourteen months after the Bentall procedure, the patient presented with sudden dyspnea and was transferred to another hospital. Upon suffering acute heart failure, the patient required mechanical ventilation and was transferred to our hospital for intubation. The patient subsequently developed severe pneumonia. As a result of prolonged ventilation, the patient underwent tracheostomy and was administered antibiotic medication (piperacillin/tazobactam) for pneumonia. Echocardiography revealed severe aortic regurgitation caused by SVD. There was a risk that a full sternotomy in a patient with tracheostomy could cause mediastinitis; therefore, we performed a re-AVR through a lower hemisternotomy (second T incision). The re-AVR surgery proceeded without complications, and the bioprosthetic valve was removed while preserving the vascular graft from the previous Bentall procedure. The postoperative course was uneventful, and the patient was discharged from hospital 31 days after the tracheostomy was closed. The success of this procedure demonstrates the viability of re-AVR through a lower hemisternotomy in patients with SVD who are at risk of additional surgical complications.
Patients with tracheostomy undergoing a full sternotomy may be at increased risk of mediastinitis and sternal infection because of possible communication between the tracheostomy and the operative field.1,2 In recent years, some clinicians have used various partial sternotomy approaches during cardiac surgery for patients with tracheostomy to avoid such complications.3,4 However, in redo cardiac surgery cases, full sternotomy is often selected to achieve adequate operative exposure to deal with severe adhesions caused by the previous operation; moreover, there are few reports of redo cardiac surgery for patients with tracheostomy. This case report presents a reoperative aortic valve replacement (re-AVR) in a patient with tracheostomy through a lower hemisternotomy after a previous Bentall procedure.
A 71-year-old man was treated using the Bentall procedure with a bioprosthetic valve (Trifecta 23 mm, Abbott, Abbott Park, IL, USA) and partial aortic arch replacement for enlargement of an ascending aortic aneurysm. He also underwent coronary artery bypass grafting with a saphenous vein graft for stenosis of the right coronary artery. Comorbidities included chronic renal failure requiring hemodialysis. Fourteen months after the procedure, the patient presented with sudden dyspnea and was transferred to the hospital where he was previously treated. Upon suffering acute heart failure, the patient required mechanical ventilation and was transferred to our hospital for intubation. Laboratory testing revealed mild thrombocytopenia (88 × 103 cells/µL) and elevated levels of creatinine (5.84 mg/dL) and brain natriuretic peptide (834.2 pg/mL). A blood culture was negative, indicating that the graft and bioprosthetic valve were not infected. Radiography showed a cardiothoracic ratio of 54%. After being transferred, the patient developed unexplained severe pneumonia; he was subsequently administered antibiotic medication and underwent tracheostomy as a result of prolonged ventilation. Transthoracic echocardiography revealed aortic regurgitation (AR) (Fig. 1A, B). The pressure half-time was shortened to 169 ms, and the patient was diagnosed with severe AR. Transesophageal echocardiography (TEE) (Fig. 1C, D) showed abnormal movement of the left coronary cusp (LCC) on the short-axis image, and AR was observed from the same site. The location of the tracheostoma is shown in Figure 1E. The saphenous vein graft had occluded before the redo surgery (Fig. 1F).
Preoperative findings.
(A, B) Transthoracic echocardiography showed aortic regurgitation (AR) and the pressure half-time was 169 ms. (C) Transesophageal echocardiography (TEE) in systole showed the abnormal movement of the left coronary cusp (arrow). (D) TEE in diastole showed AR. (E) Chest radiograph showed the tracheotoma (arrowhead). (F) Image of the aorta by three-dimensional computed tomography.
Based on these findings, the patient was diagnosed with AR and heart failure as outcomes of structural valve deterioration (SVD) and was transferred to our department for surgery. We decided to perform an urgent re-AVR. Although the patient’s severe pneumonia had improved, he needed continuous ventilation for heart failure. He underwent re-AVR with the tracheostomy in place because there was insufficient time to close it before the operation. Because of the tracheostomy, we operated through a lower hemisternotomy with an additional transverse sternum incision at the second intercostal level, described herein as the “second T incision” (Fig. 2A, B). Cardiopulmonary bypass via the femoral artery and vein was established, the ascending aortic graft was cross-clamped, and a cardiac arrest was induced with a combination of antegrade and retrograde cardioplegia. With adequate access through the second T incision, only the bioprosthetic valve was successfully removed while preserving the vascular graft from the previous Bentall procedure. We found a tear in the commissure between the LCC and the non-coronary cusp (NCC) (Fig. 2C, D), which was consistent with the preoperative diagnosis of SVD. We performed surgical aortic valve replacement (SAVR) with a new bioprosthetic valve (Inspiris resilia 23 mm; Edwards Lifesciences, Irvine, CA, USA). During the operation, the tube in the tracheostoma was temporarily changed to oral intubation, the stoma was closed with a stoma cap, and the site was covered with a drape. The times of the total operation, cardiopulmonary bypass, and aortic cross-clamp were 560, 205, and 145 min, respectively.
Intraoperative findings.
(A) A T-shaped sternotomy was performed, as indicated by the blue line (second T incision). (B) The red line shows the position of the tracheostomy. (C) Structural valve deterioration of the previous surgical bioprosthetic valve. (D) A tear of the commissure between the left coronary cusp (LCC) and the non-coronary cusp (NCC).
The patient’s postoperative course was uneventful. Ventilation was withdrawn and the patient left the intensive care unit on postoperative day 5. The tracheostomy was closed on day 21, and the patient was discharged from the hospital on day 52.
Approval from the Ethics Committee was not required because the information was analyzed retrospectively and had no effect on treatment. Written informed consent was obtained from the patient to publish this case report and the accompanying images.
In this case, the patient had severe symptomatic AR because of SVD and required emergency treatment. Although there are several reports5 of early SVD of the Trifecta valve, the underlying cause of failure has not yet been elucidated. Contact of manually tied knots with the pericardium over time could lead to leaflet tears, which could be one of the reasons for early SVD.5
For symptomatic AR, we performed re-AVR only and did not remove the vascular graft because it was not infected. The non-infected status of the graft was confirmed preoperatively using blood culture tests and echocardiography. The diagnostic distinction between non-infectious and infectious conditions is important because they may need to be treated using different approaches. If the underlying cause of AR was an infection, we would have had to remove the infected artificial graft and valve and reperform the Bentall and partial aortic arch replacement procedures; this would likely have been difficult to perform through a partial sternotomy. Furthermore, the first Bentall procedure was performed using a skirted technique, which involves the use of a composite graft created by sewing a bioprosthetic valve into the graft in advance. The graft’s lower part (skirted portion) is then sewn to the aortic annulus. In brief, the aortic annulus, graft, and bioprosthetic valve were not sewn with a continuous thread. This was also important for re-AVR because we needed to remove only the surgical valve from the aorta. We did not reconstruct the saphenous vein graft during the redo surgery even though it had become occluded because this occlusion might not have been the cause of heart failure, which was the main symptom of this patient.
For cardiac surgery, there are various alternative approaches to conventional full sternotomy, such as partial sternotomy and intercostal incision.6,7 Although a full sternotomy is still standard procedure for redo cardiac surgery, this approach increases the risk of mediastinitis and sternal infection in a patient with tracheostomy. Indeed, some reports on increased perioperative mortality in such scenarios can be found.2,3,8 Therefore, after using three-dimensional computed tomography (Fig. 1F) to plan the operative strategy, we decided to perform re-AVR through a second T incision with an adequate operating field with clamping of the ascending aorta. At least four further advantages exist in the use of a second T incision: skin and sternal incisions can be small,3,6 there is no requirement for special surgical instruments,6 the method provides adequate access to the aortic valve,8,9 and the procedure can be converted to a full sternotomy if necessary.9 The J incision, which has only one-side transverse sternum incision, could be optimal to ensure the sternum’s stability. However, we chose not to use the J incision because of the inadequate surgical field for the redo AVR after the Bentall procedure. Furthermore, although the occluded saphenous vein graft was not re-bypassed, had we found it necessary, the T incision would have enabled re-bypassing, whereas use of the J incision would have excluded this option.
As an alternative to redoing an SAVR, valve-in-valve transcatheter aortic valve replacement (ViV-TAVR) could be used for high-risk surgical patients because it has a lower 30-day mortality rate than redo-SAVR.10 Although the valves used for ViV-TAVR were not covered by insurance for a patient undergoing hemodialysis in Japan at the time, some devices have been covered since 2021. We shall consider improved therapeutic options if we face a similar case in the future. For the present case, after considering the size of the aortic root as measured by enhanced computed tomography, we considered options to implant a 20-mm Sapien 3 valve (Edwards Lifesciences, Burgess Hill, UK) or a 26-mm Evolut PRO+ valve (Medtronic, Dublin, Ireland) via the right femoral artery. However, ViV-TAVR procedure with the Sapien 3 is contraindicated for the Trifecta valve because of the high risk of coronary obstruction, and the Evolut PRO+ is not covered for patients undergoing hemodialysis in Japan. Therefore, we had no choice but to use SAVR for this patient. Of course, ViV-TAVR will become an option in the future, and accurate examination and careful selection of surgical valves with consideration of future ViV-TAVR will be important. Establishing the ViV procedure and understanding the risk of complications are important because early SVD of the Trifecta valve is commonly troublesome, as in this case. Coronary occlusion can be one of the most dangerous complications and should be avoided when possible. Occlusion occurs when TAVR is used to push the torn leaflet out to the right coronary cusp (RCC) or the LCC. The coronary entrance will not become occluded if the SVD is on the NCC but it can occur if the damage is on the RCC or LCC. To estimate the risk of occlusion, we should evaluate dimensions around the valve structure by computed tomography and using the 3mensio Structural Heart application (Photron M&E Solutions, Tokyo, Japan). In the current patient, we measured the distances from the bottom of the Valsalva to the right and left coronary entrances, which were 11.7 mm and 10.3 mm, respectively. These dimensions showed that the coronary entrance level was high enough that the leaflet could not reach sufficiently far to cover the entrance, even if the leaflet was pushed out to the cusp. We also measured the diameter and area of the aortic annulus, but these measurements indicated normal TAVR. The intraoperative findings showed the leaflet tear in the commissure between the LCC and NCC, which was the same finding by preoperative TEE. The risk of coronary occlusion was not adequately evaluated during the surgery because of the previously indicated inability to have the valve-in-valve procedure. However, ViV-TAVR could be an option in this scenario based on an evaluation of the results from 3mentio.
In summary, we successfully performed re-AVR for SVD through a second T incision in a patient with tracheostomy after a previous Bentall procedure. The patient showed no postoperative complications and was discharged on postoperative day 52.
We thank Sosuke Myojin and Kentaro Hayashida (Department of Cardiology, Keio University School of Medicine, Tokyo, Japan) for their advice. We also thank Editage (www.editage.jp) for English language editing. No funding was received for this work.
The authors declare no conflict of interest.
