Prognostic impact of hospital readmissions after primary percutaneous coronary intervention

For all patients with STEMI (from January 2003 to June 2009) undergoing primary PCI enrolled in the REAL registry, we evaluated the incidence of death and hospital readmission due to cardiovascular causes up to 3 years. Particularly, myocardial infarction (MI), coronary revascularization (CR), acute or congestive heart failure (HF), and serious bleeding events (SBEs) were assessed. An SBE was defined as hemorrhagic stroke or as any bleeding leading to hospitalization and/or transfusion of red blood cells.

The end points were assessed on December 31, 2009. The cumulative probability of adverse events was determined by the Kaplan-Meier method. Statistically significant differences between curves were evaluated with the log-rank test. To investigate the association of hospital readmission with the incidence and timing of mortality, Cox models were fitted with each adverse event as a time-updated binary covariate. These models were fitted with an adjustment for the baseline variables. P < .05 was considered statistically significant. All analyses were performed with the SAS 9.1 system (SAS Institute Inc, Cary, North Carolina).

The study population included 11118 patients. Overall, we observed 1779 deaths. The cumulative incidence of death was 12.1% at 1 year and 17.5% at 3 years. After the index hospitalization, we observed 7867 readmissions. The most frequent diseases or conditions associated with hospital readmissions were CR (28%), acute or congestive HF (26%), supraventricular arrhythmias (SA) (13%), MI (11%), and SBE (10%). Interestingly, the patients with 1 hospital admission for HF, SA, or SBE frequently showed a second (or more) readmission for the same cause (36% [95% CI, 33%-39%], 34% [95% CI, 30%-38%], and 22% [95% CI, 19%-26%], respectively). All these events tended to be more frequent during the first year.(This was particularly evident for CR. Approximately 60% of these procedures was target vessel revascularization, whereas approximately 17% was staged intervention. The influence of these readmissions on mortality varies consistently. Considering SA, no influence on mortality was found. Hemorrhagic stroke showed the strongest association with death. Coronary revascularization showed a low impact on mortality, although still numerically relevant and statistically significant. After a recurrence of MI, a quarter of the patients died, whereas almost half died after a hospital readmission for HF or SBE.