Background: The use of the methodology of adjusted risk to check the calculation of the differences between patients, groups and populations, as regards outcomes, is being used more often in the era of modern interventional cardiology to validate a reliable and balanced comparison of results between institutions and to maintain uniformity of data, criteria and definitions with a view to carrying out multicenter studies in this area of medicine. In this way, the ability to determine which variables have greater predictive value for adverse events that result from percutaneous coronary intervention (PCI) is an extremely important tool for clinical decisions and for risk adjustment in each patient, when evaluating the quality of the health care given.
Methods and results: this work took the research design of a case-control study. The data analyzed related to all patients who had undergone PCI in an interventional cardiology unit, during 2002 (567 patients). The group of cases (33) corresponds to those patients in whom a major adverse cardiac or cerebrovascular event (MACCE) occurred during the procedure or within a 30-day period. The control group (534 patients) was made up of the rest of the population that underwent a PCI but that remained free of events in the same study period. The data on the PCIs carried out was taken from the database of the interventional cardiology unit. Analysis of the data was based on a descriptive and analytical statistical treatment (SPSS 11.5) with recourse to contingency tables, the Pearson's chi-square test, the Fisher exact test, and simple and multiple logistic regression, using odds ratios (OR) as a measure of association. The level of significance was 0.05 and the confidence interval was 95%. It should be noted that in the group of variables that showed greatest predictive value of the occurrence of an MACCE we found the following: female gender (OR = 1.593), advanced age (80 years, OR = 9.460), diabetes (OR = 3.063), chronic renal failure (OR = 3.063), depressed ejection fraction (30%, OR = 8.475), priority at PCI (OR = 7.108), multivessel disease (OR = 1.683), type C lesion (OR = 2.208), acute myocardial infarction (OR = 5.045) and cardiogenic shock (OR = 28.169).
Conclusions: This study allowed us to identify, characterize and grade the set of variables that presented the strongest association with major adverse events that resulted from PCIs, for the population that underwent this type of procedure in an interventional coronary unit over the course of one year. At the same time, it enabled us to characterize the procedures that were carried out during this period, and also the demographic and clinical profile of the population. Although in some of the variables the OR values found were not statistically significant in either bivariate or multivariate analysis, we should point out that these variables are comparable to most results from similar multicenter international studies carried out with large populations. This type of study is an important contribution to improving the quality of health care given and to more efficient risk management in the field of interventional cardiology.