Short- and long-term outcomes in octogenarians after coronary artery bypass surgery

Abstract

OBJECTIVES

Coronary artery surgery in octogenarians is carried out with an increasing frequency. We tried to determine short- and long-term outcomes and quality of life after coronary artery surgery in this patient group.

METHODS

From 3312 patients undergoing isolated coronary artery bypass graft (CABG) surgery in two centres in the years 2004–06, 240 (7.2%) were older than 80 years (mean age 82.3 years, 57.1% male). The octogenarians were analysed regarding perioperative major adverse cardiac and cerebrovascular events (MACCE), late mortality and health-related quality of life (SF-12 questionnaire) and compared with 376 younger patients (mean age 66.8 years, 61.4% male) using propensity score matching. The mean follow-up time of 30-day survivors was 53 months, and follow-up completeness was 97.1%.

RESULTS

The octogenarians' 30-day mortality rate was 6.8% (vs 1.6% in the younger group). In the multivariate analysis, age was a risk factor for early death [odds ratio (OR) 4.28, 95% confidence interval (CI): 1.59–11.53] and perioperative MACCE (OR 2.78, 95% CI:1.44–5.37). One-year and 3-year survivals were 94.5 and 81.4% in the octogenarians and 98 and 91.3% in the younger group. Four years after surgery, 95.2% of the octogenarians lived alone, with a partner or with relatives, and only 4.0% required permanent nursing care. 83.9% of the octogenarians would recommend surgery to their friends and relatives for relief of symptoms.

CONCLUSIONS

Octogenarians can undergo CABG surgery with an acceptable risk of early death. Though late mortality is high, late quality of life is comparable with that of younger patients.

INTRODUCTION

In Germany, the proportion of persons aged 80 years and more is expected to increase from 8% in 2008 to 14% in 2060 [1]. The increase in life expectancy subsequently leads to a large number of patients with advanced age. Surgical treatment of coronary artery disease in the elderly involves a higher risk of morbidity and mortality than that of the younger patient group [2–7]. However, several studies documented an improved quality of life (QoL) in octogenarians after cardiac surgical procedures [8–10].

The aim of our study was first, to compare the incidence of postoperative morbidity and mortality among octogenarians vs patients <80 years who underwent coronary artery bypass grafting and second, to assess the health-related QoL after surgery.

PATIENTS AND METHODS

From January 2004 to December 2006, 3312 patients at our two centres underwent isolated coronary bypass grafting. In a first step, those who were at least 80 years and older (n = 240, 7.2%) were selected for this study and compared with younger patients (n = 3072, 92.8%). In the next step, we used propensity score matching to create two groups with comparable baseline conditions.

After this step (Fig. 1), we compared 220 octogenarians with 376 propensity score-matched patients aged <80 years with respect to their early outcome (major adverse cardiac and cerebrovascular events, MACCE), and 124 octogenarians with 280 younger patients regarding late mortality and QoL.

MACCE was defined as 30-day mortality (death from any cause within 30 days after the operation), myocardial infarction (appearance of new Q waves on the postoperative electrocardiogram, increase of blood levels of biochemical markers, new wall motion abnormalities evidenced by echocardiography), renal failure (a new need for dialysis) and neurological complications (transitory ischaemic attack, stroke, seizure, coma) after surgery.

In this study, we used a shorter version of the SF-36 (SF-12 health survey), because it is a self-report questionnaire with only 12 questions and added questions regarding medical attendance and life conditions. The SF-12 health survey includes two questions concerning physical functioning, two on role limitations because of physical health problems, one on bodily pain, one on general health perceptions, one on vitality (energy/fatigue), one on social functioning, two on role limitations because of emotional problems and two on general mental health (psychological distress and psychological well-being).

We could evaluate 404 questionnaires (Fig. 1) from 124 patients in the octogenarian group and from 280 patients in the younger group.

A summary of preoperative patient characteristics is presented in Table 1. Risk factors and concomitant diseases are perfectly well balanced between the groups, showing that the propensity score matching worked well. The fact that the EuroSCORE 1 is significantly higher in the octogenarian group does not speak against this: Every 5 years of age adds one EuroSCORE point, therefore the octogenarian group must have a higher EuroSCORE 1 risk than the younger group. The EuroSCORE 1 imbalance, thus, comes from the different ages in the investigated groups. Obviously, we could not match for age.

Statistical methods

Statistical analyses were performed using the statistical program R, a free available program (http://www.r-project.org), version 2.8.1. Statistical comparisons were performed using Fisher's exact test for parameters with normal distribution and Mann–Whitney–Wilcoxon test for not normally distributed parameters. Variables potentially associated with MACCE known from the univariate analysis were assessed by multiple logistic regression analysis and their odds ratios (ORs) are reported with 95% confidence intervals (CIs). The C-index and the Hosmer–Lemeshow goodness-of-fit test were calculated for the logistic regression model.

The propensity score analysis was carried out considering the following preoperative risk factors: EuroSCORE, diabetes mellitus type II, arterial hypertension, hyperlipoproteinaemia, haematocrit.

Differences were considered significant at a P-value of 0.05 or less. To describe survival characteristics, we used the Kaplan–Meier method.

RESULTS

With propensity score matching, we attempted to provide an unbiased estimation of treatment effects (Table 1) and compared the data of 220 octogenarians with that of 376 patients in the younger group. Consequently, there were no differences between the two groups regarding the preoperative characteristics.

Though the number of grafts was similar in both groups (Table 2), mean aortic cross-clamping time and mean operative time were significantly higher in the younger group due to the higher percentage of arterial revascularization. Off-pump revascularization was rarely (1.8% octogenarian group, 0.5% younger group) carried out in both groups.

Early results

The incidence of postoperative complications was higher in the octogenarian group (Table 2): Twenty-nine patients (13.2%) of the octogenarians required prolonged ventilatory support with intubation exceeding 24 h compared with 18 patients (4.8%) of the younger group (P ≤ 0.003). Reintubation because of respiratory insufficiency was more frequent in the octogenarian group (P = 0.004). Therefore, the length of stay in the intensive care unit was significantly (P = 0.005) longer in the octogenarian group.

Postoperative cerebral complications occurred significantly (P = 0.008) more often in octogenarians: Twelve patients in the octogenarian group (5.5%) and 6 in the younger group (1.6%) suffered from cerebral complications. The 30-day mortality rate (6.8% in the octogenarian, 1.6% in the younger group) was significantly (P = 0.004) higher for octogenarians. In the octogenarian group, six patients died within 30 days from cardiac causes (cardiogenic shock, ventricular arrhythmias) and nine died from non-cardiac causes (multiorgan failure, mesenteric ischaemia or sepsis). In the younger group, five patients died within 30 days from cardiogenic shock, ventricular arrhythmias and one died from peritonitis.

In our multivariate analysis, we could identify older age (OR 2.78; 95% CI: 1.44–5.37) and renal function (OR 2.44; 95% CI:1.29–4.63) as risk factors for MACCE, using diabetes mellitus, haematocrit and EuroSCORE as confounders (Table 3).

Late results

The mean follow-up time of 30-day survivors was 53 months (range: 44–79 month) for the whole group of 404 patients (see Fig. 1), follow-up completeness for octogenarians was 97.1%, and for the younger group it was 97.3%.

Twelve-month and 36-month survivals were 94.5 and 80.4% in the octogenarians and 98 and 91.7% in the younger group (Fig. 2).

Ninety-three patients died during the follow-up time: 24.9% (n = 51, 11 cardiac deaths, 31 non-cardiac deaths, 5 combined death causes, 5 unclear reasons for death) of the patients in the octogenarian group vs 11.4% (n = 42, 10 cardiac deaths, 32 non-cardiac deaths) of the patients in the younger group. No reoperations, but three percutaneous coronary interventions (PCIs) in the octogernarian group and six PCIs in the younger group were performed after discharge from hospital.

95.2% of the octogenarians lived alone, with a partner or with relatives; only 4.0% required permanent nursing care. 83.9% of the old patients would recommend surgery for relief of symptoms to friends and relatives and 87.9% estimated their postoperative condition to be better than the preoperative (Table. 4).

In our study, scores for the mental component were better than scores for the physical component (Fig. 3). The summary scores for QoL (Fig. 3) showed slightly worse values for the octogenarian group. In both groups, mental health had the highest score, social functioning was in general good. Role limitations caused by physical health problems had the worst scores in both groups (Fig. 4).

DISCUSSION

Our study shows that cardiac surgery in octogenarians can be performed with acceptable perioperative results: the 30-day mortality in our patient cohort is 6.8% and the long-term survival (median FU 4.4 years) 74.9%. These results of our study are in concordance with reported results: Mortality rates are reported to be between 7.1 and 8.8% for 30 days, long-term survival ranges between 60.4 and 66% after 5 years [11–13].

In our study, age itself was a risk factor for early death (OR 4.28) and perioperative MACCE (OR 2.78).

Nevertheless, octogenarians admitted for coronary surgery are not very much interested in life prolongation, but in the improvement of their QoL. We could show that after surgery, QoL is similarly good in octogenarians and in younger patients. Nearly 90% of the old patients would recommend surgery for relief of symptoms to friends and relatives and estimated their postoperative condition to be better than the preoperative.

Our results are in line with the work of Krane et al. [11], Kurlansky et al. [12] and Folkmann et al. [14], who found that the QoL in octogenarians after cardiac surgery was equivalent to or even better than that of the general population in the same age. Chaturverdi et al. [15] describe that octogenarians after surgery remain at home and function independently. Their reported rate of patients in nursing care (4.3% after 3.6 years) was very similar to ours: 95.2% of the octogenarians lived alone, with a partner or with relatives; only 4.0% required permanent nursing care. Even after non-elective surgery, octogenarians can have a QoL that equals that of the general elderly population [16].

The biggest threat to patients undergoing cardiac surgery is the risk of stroke. While octogenarians have a significantly higher (OR = 3.4) risk of stroke than younger patients (Table 2), the early mortality after a neurological event is similar in both groups: 3/12 octogenarians and 2/6 younger patients died after stroke in the early postoperative phase. During the long-term observation period, however, stroke occurred comparably often in the octogernarian (4.0%) and in the younger group (5.4%), (Table 4).

The fact that the majority of octogenarians (9/15) died from non-cardiac causes, while only 1 of 6 early deaths in the younger group was due to a non-cardiac event, indicates that ocotgenarians possibly tolerate infections, renal insufficiency and re-intubations less than younger patients. We have suggested before [17], that frailty and age-related cellular and metabolic factors may lead to this lower complication tolerance.

Interestingly, with increasing time distance from surgery, non-cardiac deaths occur similarly often in octogenarians (70.6%) and in younger patients (76.2%). This underlines the suggestion that the lower perioperative complication tolerance accounts for the majority of deaths in octogenarians. While there is an ongoing discussion whether minimally invasive surgery like off-pump coronary artery bypass surgery or transcatheter aortic valve replacement can reduce the incidence of perioperative non-cardiac complications, special perioperative care programmes for elderly people are very uncommon. Whether such a programme could be effective in avoiding perioperative complications, thus lowering the early death rate, is an interesting question.

The German health system offers the possibility of cardiac rehabilitation to every patient directly after cardiac surgery, which makes a comparison of our data with those of Bardakci et al. [18] very difficult: The authors describe that discharge to home rates for octogenarians are significantly lower than for younger patients . Because surgery is done to ameliorate long-term QoL for coronary artery bypass graft (CABG) patients, a ‘discharge to home rate’ is only useful to assess the influence of cardiac surgery on short-term QoL.

Interestingly, the in-hospital outcomes of percutaneous coronary interventions in patients >85 years compared with younger patients [19] are very similar to the surgical outcomes described in this paper: Mortality after intervention was 6.9% in the old and 1.2% in the young group. Appleby et al. [19] too, found that age was a predictor of mortality in octogenarians. It would be good to know whether in octogenarians, coronary intervention and coronary surgery would have similar results. Quite often, patients are denied coronary surgery because of their age alone, stating that these patients would never be able to stand on their own legs after surgery.

Another reason for not admitting octogenarians to cardiac surgery is the missing cost-effectiveness for a country's health system, ‘it́s not worth the effort anymore’. Expressed as cost/QoL-adjusted life year (QUALY) by Gelsomino et al. [20], octogenarians showed a much worse cost-effectiveness ($1391/QUALY) than younger patients ($516/QUALY) [10]. The cost-effectiveness ratio for heart surgery in octogenarians was very high ($94 426), compared for example, with medical therapy, which brings worse clinical results, but is much cheaper.

LIMITATIONS

Not all patients could be relocated in the follow-up period. Also, it was impossible to investigate all causes of death. However, our follow-up of 97.1% in the octogenarian group and 97.3% in the control group was sufficient to make scientific statements.

The QoL could not be assessed in all patients, because a few refused to participate in the study. However, out of 205 questionnaires sent to octogenarians, we received 194 responses.

We conclude from our results that octogenarians can undergo CABG surgery with an acceptable risk of early death. Though late mortality is high, late QoL is comparable with that of younger patients.

Conflict of interest: none declared.

REFERENCES