Objective: There has been a gradual increase in the number of elderly patients referred for cardiac surgery. These patients present a difficult challenge, they are usually symptomatic yet at high risk for intervention. The aim of this study is to review our experience with cardiac surgery in patients aged 80 years or older. Patients and methods: Between January 1981 and October 1997, 242 patients; 135 female, 107 male, mean age 82.8 years (range 80–95) underwent surgery on cardiopulmonary bypass in our unit. Surgery was performed on 14 as an emergency and 136 on an urgent (patient restricted to a hospital bed due to symptoms) basis. Pre-operatively 182 (75.2%) were in NYHA functional class 3 or 4. Results: Early mortality was 14 (5.7%). A mitral valve procedure and emergency surgery were significantly associated (P<0.05) with an increased risk of operative mortality. Median ITU and in-hospital stay was 1 day (range 0–33) and 10 (range 6–49) days, respectively. Ninety-three percent of patients were living independently at home 2 months post-operatively. Survival (±SEM) is 98% complete (totals 557 patient years) and including early mortality at 1 and 5 years was 85.5±2.4% (n=154), and 67.7±4.3% (n=33). Survival for patients undergoing isolated aortic valve replacement (AVR) and coronary artery bypass grafting (CABG) at 5 years was 64.8±7.8% and 79.7±7.4%, respectively. Survival was significantly worse in patients undergoing a mitral procedure. Using Cox's proportional hazards model only type of operation (mitral surgery) was significantly associated with worse survival. Conclusion: Cardiac surgery can be performed in a selected elderly population with a low operative mortality. Post-operatively elderly patients attain an excellent quality of life and survival. Emergency and mitral surgery in this group of patients is less rewarding.
The mean age of the population is increasing. Those aged over 80 are the fastest growing section of the population . In the UK in 1994, 4% of the population were aged 80 or older and by 2020 it is estimated that this figure will approach 6% . This approximates to 3 600 000 people. As many as 25–50% of these people will be limited by cardiovascular disease .
Following several encouraging reports ,,,,, there has been a gradual increase in the number of elderly patients referred for cardiac surgery. Not surprisingly, as experience has increased, the risk profile of these patients has worsened. Many of these patients are now restricted to hospital pre-operatively and a small but significant number undergo surgery as true emergencies. The expectation of both referring physicians and patients has changed and accurate documentation of outcome following surgery in these patients is necessary.
Furthermore there is no obvious relationship between the proportion of elderly in the population and the percentage of gross national product spent on health  but with increasing pressure on resources, clinicians must justify expensive interventions such as cardiac surgery in the elderly.
Thus we report our experience with cardiac surgery in octogenarians.
Patients and methods
We have retrospectively reviewed the records of all 242 patients over the age of 80 years who have undergone cardiac surgery between January 1981 and March 1998 at the Wessex Cardiothoracic Centre. There has been a progressive increase in the number of these patients (Fig. 1 ).
There were 135 females and 107 males. The mean age of the group was 82.8 years (range 80–95). The procedures performed included AVR (118), AVR+CABG (39), CABG (50), mitral valve replacement (MVR)/MVRepair (15), MVR/MVRepair+CABG (six) and others (14). Four were redo operations.
All patients underwent surgery using cardiopulmonary bypass with the preferred myocardial preservation technique of the individual surgeon.
The fitness and discharge details of each patient were obtained from the notes, the patient or their general practitioner. Survival information was obtained from general practitioners (GPs) and the Office of National Statistics. Patients dying within 30 days of surgery or dying in hospital more than 30 days after surgery were included in early mortality.
The association between variables and operative mortality was assessed using logistic regression and chi-squared tests. Kaplan–Meier survival curves are presented. Factors associated with survival were ascertained using Cox's proportional hazards method.
The NYHA functional status of these patients preoperatively was 7 class I, 53 class II, 96 class III and 86 class IV. NYHA status for the different procedures is shown in Table 1 .
The mean aortic valve gradient was 97 mmHg (range 45–161) for patients undergoing AVR and 80 mmHg (range 40–130) for patients undergoing AVR+CABG. The majority (80%) of those patients with isolated coronary disease had triple vessel disease.
Fourteen patients underwent surgery on an emergency basis, 136 urgent (patient restricted to hospital despite maximal therapy) and 92 on an elective basis (Table 2 ). There was no significant difference in the proportion of patients undergoing each operation and urgency of surgery.
In the aortic position the majority of valves implanted were stented tissue prostheses (116/118). Mechanical (2/118) valves were only implanted if another indication for anticoagulation was present. In the mitral position, where possible valves were repaired (5/21), otherwise the valves implanted were predominantly mechanical (10/21). Internal mammary arterial grafts were used in only eight patients.
Early mortality was 14 (5.7%). The only pre-operative or operative variables associated with an increased risk of operative mortality were mitral procedure (P=0.04) and urgency of surgery (P=0.05). Age, gender, functional NYHA status, number of coronaries diseased and number of bypasses (where applicable), valve gradient and size implanted (where applicable); myocardial ischemic and cardiopulmonary bypass times were not associated with an increased risk of early mortality.
The median, mean intensive therapy unit (ITU) and in-hospital stay was 1, 1.67 days (range 0–33 days) and 10, 12 days (range 6–49), respectively. Those patients who underwent a mitral procedure had a significantly longer stay in hospital (P<0.001) (Table 3 ).
Thirty-five patients remained on ITU for greater than 1 day. Reasons for the delayed discharge included prolonged inotrope/IABP requirement: (AVR, three; AVR+CABG, two; CABG, three; MVR/Rep, one; other, five), respiratory failure (AVR, four; MVR/Rep+CABG, one), resternotomy (AVR, three; AVR+CABG, one; CABG, one), CVA (AVR, three; MVR/Rep, one), prolonged inotrope/IABP requirement and respiratory failure (AVR, one; AVR+CABG, one; MVR/Rep+CABG, one; other, one) and others (AVR, two; AVR+CABG, one). All resternotomies were due to bleeding.
Fifty-nine of the 207 patients who were discharged from ITU on day 1 experienced complications post-operatively on the ward. In addition 18 patients who were on ITU for greater than 1 day, developed a complication.
For patients operated on in the 1980's it was not always possible to ascertain from the notes their level of independence and functional NYHA status post-operatively. Their GP had frequently retired or not surprisingly could not remember; it was only possible therefore to accurately determine functional NYHA status in 123/228 (54%) and ability to live independently in 179/228 (78.5%).
Ninety-three percent of patients were living independently at home (167/179) 2 months following the procedure. Seven of the 12 patients unable to live independently had non-cardiac problems (one blind, one osteoarthritis, two CVA, three respiratory disease) and five had residual severely limiting cardiac symptoms. There was an improvement in mean NYHA class for each group of patients (Fig. 2 ).
Survival (±SEM) (Fig. 3 ) including early mortality at 1 and 5 years was 85.5±2.4% (n=154) and 67.7±4.3% (n=33). The dashed line represents the expected survival of the population in the UK aged 80 years old. There was no significant difference in survival between the general population and those who underwent cardiac surgery.
Survival for the most common procedures is shown in Fig. 4 . Survival (±SEM) at 1, 3 and 5 years was 92.4±2.6% (n=86), 84.2±4% (n=46) 64.8±7.8% (n=16) for AVR, 82.1±6.7% (n=24), 63.6±9.7% (n=13), 57.2±10.6% (n=4) for AVR+CABG and 91.3±4.2% (n=34), 79.7±7.4% (n=10), 79.7±7.4% (n=2) for CABG. Patients who underwent a mitral valve procedure, demonstrated significantly worse survival than those who did not (Log Rank P<0.001, Wilcoxon P<0.001) (Fig. 4).
Sixty patients have died since their operation, 14 early and 46 late, 11 of the late deaths were cardiac related (three, AVR; one, AVR+CABG; one, CABG; four, MVR/Rep and two, others), 32 non-cardiac related and three unknown. Using Cox's proportional hazards model only the type of operation (mitral replacement or repair±CABG) significantly adversely affected survival (Table 4 ).
Accepting the difficulties and pitfalls of a retrospective analysis, this series clearly demonstrates that for selected patients greater than 80 years of age cardiac surgery can attain useful results.
Operative mortality and morbidity, ITU and in-hospital stay
Emergency surgery was significantly associated with an increased risk of operative mortality. This has been shown in previous studies in the elderly population . In our experience this group includes patients with post-infarct VSD's and acute aortic dissection, a subset of patients who do poorly in any series. We therefore do not believe that emergency status alone is a contraindication to surgery in these patients but where possible delays in surgery are best avoided . We believe there is a smaller window of opportunity in these patients compared with younger patients with similar pathologies. Unlike other authors we found no difference in operative mortality between elective and urgent cases .
The total operative mortality and mortality as per procedure compare favourably in this series with previous reported series (Table 5 ).
We believe that our figures represent modern practice. More recently, operative mortality has been compared to date of surgery and some series report similar operative mortality for operations performed from 1992 onwards. ITU and in-hospital stays are lower than previously published and represent modern practice.
A mitral valve procedure was associated with an increased operative mortality and longer in-hospital stay. Most authors report a higher incidence of operative mortality for mitral valve procedures ,,,. In our series, all deaths followed mitral valve replacement (4/18) rather than repair (0/7). It is impossible to draw far reaching conclusions from these data because of the variability in the small number of patients, however, it highlights the need for caution in recommending surgery for mitral valve disease in the very elderly especially when echocardiography suggests that the valve is not repairable. The relatively high number of tissue valves implanted in the mitral position represents our policy that tissue prosthesis are the implants of choice in the elderly, when no indication for anticoagulation is present. The improvement in mean NYHA status was greater in those patients who did not undergo a mitral procedure. This poorer functional improvement combined with the increased risk emphasizes the difficulty of mitral valve procedures in the elderly.
Unless an indication for anti-coagulation is present, the majority of surgeons favour implantation of a biological prosthesis in these patients. To comment on valve related complications when a limited number of patients receive a variety of different implants is not ideal, though no patients with biological valves have required reoperation for structural dysfunction of the implanted valve.
Survival following cardiac surgery in patients aged 80 years or older is the same as that of the general population aged 80 years or older. This suggests that serious non-cardiac disease in these elderly patients who undergo surgery is not prevalent when compared to the elderly population at large. This probably reflects careful selection by referring physicians. The same contraindications for surgery are used as for a younger population.
As in any study on a number of different cardiac procedures, survival was significantly worse in patients who underwent a mitral procedure compared to an isolated aortic or bypass, or aortic and bypass procedure. Differences due to small sample size probably explain the discrepancy between survival in this series, 33% at 5 years for any patient undergoing a mitral procedure when compared to a recent reported analysis from the UK heart valve registry, 40% at 5 years .
One hundred and thirty-six patients underwent surgery on an urgent basis. These patients were in hospital at the time of referral and due to refractory symptoms despite maximal medical therapy, were unable to be discharged prior to surgery.
Ninety two percent of these urgent patients were able to live at home independently 2 months after surgery. Treated medically, the majority of these patients would have remained in-patients. It is impossible to compare the cost of medical treatment versus surgery in this situation, but given these results a beneficial role of surgery is clearly demonstrated.
We believe that elderly patients with isolated, severely limiting cardiac disease can benefit from cardiac surgery with an acceptable operative mortality. Survival following the surgery is similar to that of any elderly patient. Patients undergoing mitral valve procedures need particularly careful evaluation.
- coronary artery bypass surgery
- cardiopulmonary bypass
- cardiac surgery procedures
- mitral valve
- aortic valve replacement
- aged adult, 80 and over
- intensive care unit
- surgical procedures, operative
- quality of life
- surgery specialty
- independent living
- new york heart association classification
- surgical mortality
- emergency surgical procedure