Factors predictive for in-hospital mortality following percutaneous coronary intervention

Background. - Despite advances in procedures for percutaneous coronary intervention (PCI) and enhancement of materials and adjunctive therapy, postprocedural mortality remains a possible adverse outcome after PCI. Aims. - To assess factors independently associated with in-hospital mortality in patients referred for PCI. Methods. - Between January 2004 and December 2005, 4074 PCI were performed in our University Hospital, with 70 deaths registered either during the procedure or during the in-hospital stay. The 70 patients who died were age- and sex-matched with 70 controls in a case-control design study. Clinical and angiographic characteristics at hospital admission were collected from the patients' medical files. Results. - The cumulative incidence rate for in-hospital mortality was 1.72%. Variables positively and significantly associated with in-hospital mortality were severe renal failure (55.7% in cases versus 12.9% in controls, p < 0.0001), cardiac failure (26.1% versus 10.1%, p = 0.01), ST-segment elevation myocardial infarction (STEMI) (70.6% versus 31.4%, p < 0.0001), proximal coronary lesion (72.9% versus 40.0%, p < 0.0001) and angiographically visible thrombus (14.3% versus 4.3%, p=0.04). Conversely, history of coronary heart disease, smoking and dyslipidemia were less frequent among cases. In multivariable analysis, the adjusted odds ratios (OR) for in-hospital death were 4.89 (95% confidence interval [CI] 1.96-12.2, p < 0.001) in STEMI versus non-STEMI, 4.28 (95% CI 1.73-10.6, p < 0.01) in those with a proximal coronary lesion, and 9.77 (95% CI 3.42-27.9, p < 0.0001) in patients with severe renal failure. Conclusion. - STEMI, proximal coronary lesion, and renal failure at admission are identified as particular settings associated with a higher probability of in-hospital mortality after PCI. (C) 2008 Elsevier Masson SAS. All rights reserved.