Discussion
The present study tries to address the important question whether it is still reasonable to undertake a re-sternotomy for AVR in patients with patent CABG in the era of transcatheter valve therapy.
The reported mortality of re-sternotomy is 6-20% (5-7). Mayo Clinic reported 8% early mortality for conventional redo biological valve replacements over 20 years in a mixed series of aortic and mitral valve replacements. New York Heart Association functional class (hazard ratio, 2.1; 95% confidence interval, 1.06-4.3; P = 0.03) and prior CABG (hazard ratio, 3.5; 95% confidence interval, 1.2-10.9; P = 0.03) were independent predictors of early death (8). Survival at 5 and 10 years was 63% and 34% respectively. Patients with the combination of prior CABG and New York Heart Association functional class III or IV accounted for 46% of early deaths.
Our in-hospital mortality of 7.8% compares favourably with these results. We divided our patients into 2 subgroups based on the premise that group 1 (isolated AVR with re-sternotomy and no other concomitant cardiac procedure) have the option of TAVI. They can potentially avoid a risky and complex operation and have a reduced hospital stay and quicker recovery.
The in-hospital mortality in group 2 was almost twice that of group 1 (10.2%, group 2 versus 5%, group 1, p=0.247). Although, this did not reach statistical significance, a concomitant procedure does add to the complexity of surgery and increases risk.
These mortality rates are still remarkably high compared to isolated first time AVR with mortality of 0.5-1.9% (9-12). In addition to the hazard of surgery itself, there may also be the additional hazard associated with long standing ischemic heart disease in combination with aortic valve disease. Our analysis showed that the long-term survival of patients undergoing first time isolated AVR approaches that of an age-matched general population whereas it lags behind significantly after a re-sternotomy AVR in patients with previous surgical myocardial revascularisation.
Attempts to dissect and control LIMA increases risk of injury and operative mortality (13,14). Patent LIMA in a perfused heart bloodies the operative field with inconvenience of stitching on a beating heart. In conjunction with moderate to deep hypothermia on a fibrillating heart, this technique has been considered safe and avoids injury to LIMA (15,16). Systemic hyperkalaemia with adjunctive hypothermia, for diastolic arrest is a safe proposition, to avoid injury to LIMA when dissection is difficult due to adhesions.
Pre-sternotomy institution of cardiopulmonary bypass through peripheral cannulation is considered a safe strategy to manage any catastrophic bleeding. It decreases myocardial injury and complication rates, blood and blood product usage, hospital stay and hospital costs (17). In this analysis, our numbers may have been too small for estimation of the effects of LIMA injury and pre-sternotomy institution of cardiopulmonary bypass.
TAVI now provides a safe option for isolated AVR in patients with a hostile chest. These include previous multiple sternotomies, sternal infections and mediastinitis, prior irradiation, elderly, frail patients, calcified aorta/root. TAVI allows a shorter hospital stay and a lower postoperative morbidity rate compared to re-sternotomy (18,19). Stortecky et al reported that in elderly, high-risk patients after prior CABG, conventional AVR and TAVI are comparable treatment options. All-cause mortality was 2.5% in both groups and major adverse cardiac and cerebrovascular event rates were comparable (7.5% TAVI vs 17.5% S-AVR, p = 0.311) after 30 days. TAVI had a higher rate of permanent pacemaker implantation (30% vs 0%, p < 0.001) and grade II residual aortic regurgitation in 14%. Incidence of cerebrovascular events was 7.5% in SAVR vs 2.5% in TAVI (p = 0.61).
Patent grafts can also provide protection in cases of low coronary ostial heights especially for high profile transcatheter valves. There remain concerns for high paravalvular leak rates, permanent pacemaker implantation, sub-valvular leaflet thrombosis, need for anticoagulation and long-term durability of transcatheter valves (20-22).
There is a significant economic burden with re-sternotomies due to longer ventilation times, intensive care and hospital stay and need more intensive tests and treatments like inotropic support, hemofiltration, invasive and non-invasive ventilator support and intra-aortic balloon pumps. Our analysis did not evaluate the cost burden of these procedures. No direct cost comparisons are available between re-sternotomies and TAVI. With increasing TAVI volumes, greater operator experience and decreasing costs, TAVI remains an attractive option in this group of difficult surgical patients.
Best practice recommendations now include discussions in the Heart Valve Team to ensure the most appropriate procedure in terms of technical feasibility, safety and durability. Frail and comorbid patients may benefit from a less invasive intervention, however, in case of concomitant diseases, a complete treatment with a complex conventional surgical operation should not be avoided despite relatively high initial comorbidity.