Editorial analysis: impact of perfusion strategy on stroke risk for minimally invasive cardiac surgery

Abstract

The move from full median sternotomy to less invasive approaches for valvular heart surgery has been driven by the desire to reduce post-operative pain and surgical trauma and to improve cosmesis and patient satisfaction [1]. Non-sternotomy incisions, however, can limit exposure to the ascending aorta. This raises the question of whether to attempt a technically more challenging ascending aortic cannulation using advanced cannulation technology/techniques, or perform retrograde arterial perfusion via a standard femoral approach [2]. A recent Society of Thoracic Surgeons (STS) database publication associated ‘less invasive mitral valve (LIMV) operation(s)’ with a nearly 2-fold increase in the risk of permanent stroke [3]. While this analysis was hampered by its equation of LIMV with femoral perfusion, it did raise the spectre that increased stroke risk was associated with a femoral perfusion strategy. Over the past 15 years, our institution has employed different perfusion techniques for our less invasive procedures. Our initial foray into right-thoracotomy incisions utilized retrograde perfusion via the ‘port access’ platform. With an informal strategy of intraoperative echocardiographic analysis of the aortic arch and the descending aorta, we avoided the use of femoral perfusion when we believed that there was significant atherosclerotic burden. The results from this approach were excellent as demonstrated by the evaluation of our first 714 minimally invasive mitral valve procedures [4]. In this cohort, where 30% of patients were >70 years of age, 15% were reoperations and 12% were multivalve operations, femoral perfusion was used in nearly 80% of patients, with a 2.9% incidence of stroke. Although satisfied with these results, we realized that there was a significant, primarily geriatric patient population which was underserved by this selective approach. As such, we developed greater facility with central aortic cannulation through a mini-thoracotomy incision until this became our ‘go-to’ approach for the majority of our minimally invasive mitral valve procedures, regardless of age. These minimally invasive incision patients have been the subject of recent reports. At the 2011 STS meeting, we presented an analysis of 3180 isolated, non-reoperative valve procedures performed at our institution between 1995 and 2007. The overall stroke rate was 2.2%, with increased stroke risk associated with an atherosclerotic aorta, cerebrovascular disease, emergent operation, ejection fraction <30% or retrograde perfusion (P < 0.05 for each), but not with incision location (P = 0.82). Additionally, the association of retrograde perfusion became insignificant when analysing patients who are 50-years old or younger [5]. These results mirror those of our previous cohort of patients undergoing reoperative mitral valve procedures, which revealed that retrograde perfusion was the only independent risk factor for stroke (odds ratio 4.4; P = 0.001) [6]. Subsequently, we presented a focused report on a more homogeneous subset of 1282 first-time, isolated mitral valve operations performed through a right anterior mini-thoracotomy over a 12-year period [7]. This homogeneity allowed us greater discriminatory power to analyse the specific patient factors associated with an increased risk of stroke. The only significant risk factor interaction for neurologic complication identified was the use of retrograde perfusion in patients with high-risk comorbidities: peripheral vascular disease, cerebrovascular disease, atherosclerotic aortas or dialysis dependence. Our current clinical practice attempts to restrict retrograde arterial perfusion to those surgical scenarios where there is a very limited central aortic access, such as patients undergoing robotic mitral valve surgery, or a hostile mediastinum. In these cases, the question remains: what preoperative evaluation is needed prior to performing retrograde perfusion? Our data suggest that retrograde perfusion remains a viable option for younger patients without vascular co-morbidities. In older patients or those with the risk factors discussed above, we currently recommend performing a computed tomography angiography of the descending aorta with distal runoff in addition to an intraoperative transoesophageal echocardiographic assessment of the thoracic aorta. Such an approach has been shown to be efficacious by Murphy et al. [8], who demonstrated a 1.6% stroke rate using retrograde perfusion in similarly screened patients undergoing robotic cardiac procedures. Minimally invasive valve surgery with antegrade perfusion has a low risk of neurological complications and has excellent outcomes. Retrograde perfusion in older patients with significant vascular co-morbidities is associated with an increased risk of stroke. The vast majority of our patients currently undergo heart valve procedures through a right anterior mini-thoracotomy with