Cardiac surgery improves survival in advanced left ventricular dysfunction: multivariate analysis of a consecutive series of 4491 patients over an 18-year period

Objectives: Risks of cardiac surgery in patients with poor [ejection fraction (EF) ≤ 30%] and very poor left ventricular (LV) function (EF ≤ 20%) may be considered high due to increased mortality. We examine our results in this cohort of patients.

Methods: Data were prospectively collected and retrospectively analysed from 4491 consecutive patients referred for cardiac surgery over 18 years (July 1993-June 2012). Univariate predictors of in-hospital postoperative mortality were analysed by the appropriate tests. Variables with P < 0.1 were entered into multivariable logistic-regression model to identify predictors of in-hospital postoperative mortality, with data presented as odds ratios; P < 0.05 was statistically significant. Data on long-term survival and cardiac-specific mortality were obtained from the UK Office for National Statistics; the date of last follow-up was 13 October 2013 for the alive patients. Univariate predictors influencing cardiac death were determined by log-rank method. Variables with P < 0.1 were entered into multivariable Cox regression model to determine independent predictors of long-term survival, with data presented as hazard ratios; P < 0.05 was statistically significant.

Results: Cardiac surgery was performed on 3890 consecutive patients (74.7% male, age 68.7 ± 8.1 years); 601 patients did not undergo surgery. Postoperative hospital mortality was 2.9% (n = 112/3890). Predictors of postoperative hospital mortality included age ≥ 70 years, female sex, hypertension, LVEF < 50%, neurological dysfunction, previous cardiac surgery, early time period 1993-1997, emergency procedures and triple procedures. All patients were followed until the date of last follow-up or date of death, with a median follow-up of 8.1 ± 7.6 years and a total follow-up of 33 208 years. There were 533 (13.7%) postoperative early and late deaths from cardiac causes. Predictors of long-term survival free from cardiac death included LVEF > 50%. Predictors of postoperative cardiac deaths in the long-term follow-up included older age, diabetes, neurological dysfunction, LVEF < 50%, non-coronary artery bypass surgery, early time period of surgery (1993-1997) and redo-cardiac surgery.

Conclusions: Cardiac surgery provides long-term survival benefit in all subsets of LV function, including advanced LV dysfunction.