European Heart Journal Advance Access published online on February 9, 2009
European Heart Journal, doi:10.1093/eurheartj/ehp008
The impact of valve surgery on short- and long-term mortality in left-sided infective endocarditis: do differences in methodological approaches explain previous conflicting results?
1 Nancy-Université, Faculté de médecine, EA4003, Nancy 54000, France
2 Inserm, CIC-EC, Nancy 54000, France
3 CHU Nancy, Epidemiologie, CO No. 34, Nancy Cedex 54035, France
4 CHU Besançon, Maladies Infectieuses et Tropicales, Besançon 25000, France
5 APHP, Hôpital Bichat Claude Bernard, Centre d'Investigation Clinique, Maladies Infectieuses et Tropicales, Paris 75018, France
6 HCL, Hôpital Cardiovasculaire et Pneumologique Louis Pradel, Chirurgie Cardiothoracique et Transplantation, Lyon-Bron 69500, France
7 HCL, Hôpital Cardiovasculaire et Pneumologique Louis Pradel, Laboratoire de Physiologie Lyon Nord, UCBL1 Inserm U886 ‘cardioprotection’, Lyon, France
8 CHU Nancy, Cardiologie, Nancy 54000, France
9 CHU Montpellier, Maladies Infectieuses et Tropicales, Montpellier 34000, France
10 CHU Pontchaillou, Maladies Infectieuses, Rennes 35000, France
11 APHP, Hôpital Bichat Claude Bernard, Cardiologie, Paris 75018, France
12 HCL, Hôpital Cardiovasculaire et Pneumologique Louis Pradel, Cardiologie, Lyon-Bron 69500, France
Received 16 June 2008; revised 27 November 2008; accepted 5 January 2008.
* Corresponding author. Tel: +33 383852163, Fax: +33 383851205, Email: f.alla{at}chu-nancy.fr
Aims: The aim of this study was to evaluate the effect of valve surgery (VS) in infective endocarditis (IE) on 5-year mortality and to evaluate whether conflicting results reported by previous studies could be due to differences in their methodological approaches.
Methods and results: Four hundred and forty-nine patients with a definite left-sided IE were selected from a prospective, population-based study. Association between VS and 5-year mortality was examined with a Cox model. To determine the impact of different methodological approaches, we also analysed the relationship between VS and mortality in our database, according to each method used in the five previous studies. Valve surgery was performed in 240 patients (53%). It was associated with an increase in short-term mortality [within the first 14 post-operative days; adjusted hazard ratio (HR), 3.69; 95% confidence interval (CI), 2.17–6.25; P < 0.0001] and a decrease in long-term mortality (adjusted HR, 0.55; 95% CI, 0.35–0.87; P = 0.01). At least 188 days of follow-up were required for VS to provide an overall survival advantage. When applying each study's method to our database, we obtained results similar to those reported.
Conclusion: Previous conflicting results appear to be related to differences in statistical methods. When using appropriate models, we found that VS was significantly associated with reduced long-term mortality.
Key Words: Infective endocarditis • Valve surgery • Mortality • Propensity analysis
See Appendix for the AEPEI study group.