Background: Previous work from our group demonstrated age to be an ind
ependent predictor of operative mortality in valvular heart surgery; 5
.6% of patients under 70 years of age ('young') die compared with 10%
of those over 70 years ('old'). These results are largely based on pat
ients who underwent heart surgery before the introduction of oxygenate
d blood cardioplegia and before the development of retrograde delivery
systems. The goal of this study was to determine if age remains a pre
dictor of operative outcome in the current era. Methods: Between Novem
ber 1989 and December 1993 we performed 423 valvular heart procedures;
300 (71%) in young and 123 (29%) in old patients. Patients were asses
sed according to age, sex, preoperative symptoms [New York Heart Assoc
iation (NYHA) class ill and IV versus I and II], timing of surgery [el
ective versus urgent funder 72 h)], associated coronary artery disease
, associated left main stenosis, concomitant bypass surgery, left vent
ricular function (ejection fraction <40% versus >40%) and surgical pro
cedure (mitral versus aortic valve surgery). Univariate (Fisher's exac
t test) and multivariate analyses (logistic regression) were used to a
ssess the effects of the above on the outcome of surgery. Results: The
mortality was 8.1% in old compared with 2.7% in young patients (P<0.0
2), and the cumulative morbidity and mortality was 34.2% in old compar
ed with 17.7% in young patients (P<0.001). Univariate analysis identif
ied age and the timing of surgery as the only predictors of survival.
Identical results were obtained after multivariate analysis using the
stepwise logistic regression model (P-age=0.01; P-timing=0.016). No ad
ditional independent predictors of outcome could be identified. Relati
ve odds ratio determination demonstrated that the combination of urgen
t surgery and advanced age increased the risk of surgery tenfold. Conc
lusions: Improvements in myocardial protection and perioperative care
may account for the slightly better results of valvular surgery (parti
cularly in young patients) in the current study when compared with our
previous work. However, age continues to be an important predictor of
mortality. In the elderly, therefore, careful patient selection and e
arly intervention to avoid-the increased risk associated with urgent s
urgery remains the most important factor influencing operative outcome
.