Limits to surgery for aortic valve stenosis

Background

The natural history of aortic stenosis (AS) has shown that in the absence of surgical management, patients will develop progressively invalidating symptoms of syncope and angina. Mortality rates from congestive heart failure at 5 and 10 years are 68% and 82% respectively (1). Indication for surgery rises when the severity of the stenosis becomes significant (valve area <1cm² or 0.6 cm²/m² body surface area) or the patient becomes symptomatic (2).

Conventional surgical AVR is a reference treatment and is performed under cardiopulmonary bypass, cardiac arrest and aortic cross-clamping. Isolated AVR carries an average 30-day mortality rate of 3.8±1.5% (3). It is the “gold standard” for symptomatic aortic stenosis and has been shown to improve outcome and survival. Intervention indications have been revised to performing corrective procedures before establishing severe myocardial damage and even prior to onset of symptoms (4). A multivariate analysis of almost 6,000 patients having AVR, showed that the 5 most important predictors of mortality were age ≥ 80 years, NYHA class ≥ III, EF ≤ 30% associated with previous MI, emergent AVR and concomitant coronary artery bypass graft (CABG) surgery (5).

1 - Limits related to age

Cardiologists are reluctant to refer elderly and high-risk patients for AVR. Age was a recurrent factor for refusing surgery in 31.8% of patients with AS from the Euro Heart Survey on Valvular Heart Disease (6) and 62% of patients with AS in another study from the USA (7). It is an important predictor of operative risk and survival in cardiac surgery. Advanced age has repeatedly and consistently been shown to be a predictor of both poor in-hospital outcome and long term survival. In a series of 6,359 patients undergoing aortic valve replacement, Hannan et al showed an incremental increase in the adjusted hazard ratio for 30-month survival from 1.57 to 2.18 to 3.96 in age ranges 65-74 y, 75-84 y and ≥85 y, respectively. After isolated AVR, the 30-month survival was 90.1% for patients age  under 75 and 86.2% for patients >75 years of age (8). A study in octogenarians with severe AS showed that AVR had significant survival benefit with 1-year, 2-year and 5-year survival rates of 87, 78 and 68%, respectively, compared with 52, 40 and 22%, respectively, in those who received no AVR (9). Elderly patients on the other hand experience increased operative mortality and also are at higher risk for valve-related events (10,11).
Nevertheless, according to published guidelines, age is not, per se, a contraindication to AVR (12,13,14). Analysis of determinants of operative mortality in relation to age showed that age is not linearly related to mortality rates after AVR (15) and that there is considerable functional improvement that follows valve replacement (16). 

2 - Limits related to comorbidity

Patients can be refused for surgery because of severe comorbidities known to be associated with poor outcome. Comorbidities may be related to concomitant cardiac diseases which can further compromise myocardial function such as poor left ventricular ejection fraction (LVEF), previous cardiac surgery, and associated coronary artery disease (CAD). Other comorbidities relating to the general condition of the patient, such as neurological dysfunction, chronic lung disease, liver cirrhosis and renal insufficiency are additional predictors of poor outcome. These patients are prone to severe postoperative complications such as infections and bleeding; and the procedure itself may further compromise vital organ function (17,18). The contribution of these factors can increase the odds ratio for operative mortality by a factor of 10.6 for emergency versus elective surgery, 4.9 for renal failure, 3.1 for NYHA class (III-IV versus I-II) and 4.3 for neurological dysfunction. Thus, it may be too late to perform elective valve replacement on patients with terminal end-organ failure of the liver (Child-Pugh class B or C cirrhosis) or lung.

3 - Limits related to the presence of concomitant coronary artery disease

CAD especially may negatively affect prognosis in patients with AS due to the presence of a concomitant cardiac pathology and impaired LV function from an ischaemic myocardium. Patients with acute myocardial infarction (AMI) <24 hours or who were haemodynamically unstable had a risk-adjusted 30-month survival of 59.6%, compared with 83.6% for patients with neither AMI <24 hours nor haemodynamic instability. The operative mortality of AVR doubles with the addition of a concomitant CABG procedure, a figure that cannot be explained solely by the simple increment in time from cross clamping and cardiopulmonary bypass. Concomitant CABG had an adjusted 30-month mortality hazard ratio of 1.26 in comparison with isolated AVR. After AVR, LV dimensions normalises more quickly in the group with isolated AVR compared to those with concomitant CABG, further suggesting that CAD has a negative impact on postoperative myocardial recovery. The operative risk in patients with CAD requiring concomitant CABG is further compounded by a significantly higher prevalence of cerebrovascular disease, peripheral vascular disease, extensive aortic atherosclerosis, diabetes and renal failure.  Nevertheless, there is a consensus about the fact that the addition of CABG to AVR slightly improves long-term survival, even in high-risk populations (19). 

4 - Limits related to contractile reserve

Delay in the management of patients with AS may give rise to certain “cardiac” profiles that are intrinsically associated with poor outcomes following AVR. Thus when hypertrophy fails to normalise wall stress, the abnormal afterload leads to reduced ventricular ejection, reducing cardiac output, adding to the heart failure syndrome (20).  This subset of patients with low gradient AS and low EF is known to be associated with poorer outcomes after AVR. A poor outcome is seen in 5–10% of all cases of severe AS (21)  and defined as patients with a mean gradient <30 mmHg (or 40 mm Hg), an aortic valve area <1 cm², and an EF<35% (or 40%). The associated myocardial dysfunction contributes to poor prognosis. Since the transvalvular gradient is small, there is a correspondingly smaller reduction in afterload and thus a smaller improvement in EF following surgery (22). AVR in this group of patients carries a poor prognosis with a reported operative mortality as high as 21% with a 50% death rate within 4 years of the procedure (23).  Although AVR is superior to medical management in terms of short-term survival, surgery is not recommended to all low-gradient, low-EF patients.
inotropic reserve as well as a large increase in aortic valve area with increased output generally considered a contraindication and patients with these are least likely to benefit from AVR . Nevertheless, in a recent international multicenter registry of low EF/low gradient AS, AVR was associated with superior survival and was advocated when mean pressure gradient was >20 mm Hg and in the absence of excessive comorbidities or severe CAD with large scarring which would be caused by extensive myocardial infarction (24). 

5 - Technical limits to surgical AVR

In addition to comorbidities, patients may present with technical difficulties and complexities which make AVR challenging to perform. This is particularly true in patients undergoing redo surgery with patent coronary artery bypass grafts, where the risk of injury to the graft during dissection can be prejudicious to myocardial vascularisation. The matter is further complicated by issues related to cardioplegia when the patent grafts are the internal thoracic arteries. Patients with previous mediastinal radiotherapy and radiation damage to the myocardium are also known to have poor outcomes. Finally, in the presence of a heavily calcified and atheromatous ascending aorta (porcelain aorta), cross clamping and aortotomy may be impossible.