We identified 21 patients in the pre-intervention period and 31 patients in the postintervention period with definitive IE who had guideline indications for surgery. The postintervention cohort was older and had more male patients; this difference was not statistically significant. No differences were noted between the groups for race, gender, or intravenous (IV) drug use (Table 2). Chi-square tests of independence were performed to assess the relationship between age and our primary outcomes. There was a significant relationship between age and the likelihood of receiving or being transferred for surgery (59.3% vs 50% vs 7.7% for 18-44 y, 45-64 y, and ≥ 65 y, respectively; χ2 [2, N = 52] = 9.67; P = .008), but not between age and mortality (14.8% vs 25.0% vs 30.8% for 18-44 y, 45-64 y, and ≥ 65 y, respectively; χ2 = 1.48 [2, N = 52; P = .478]. The electronic version of the clinical pathway was activated and used in only 3 of the 31 patients in the postintervention period. Consultations to ID, cardiology, and CTS teams were compared between the study periods (Table 2). Although more consultations were seen in the postintervention period, differences were not statistically significant.
The unadjusted primary outcomes are shown in Table 2. More surgeries were performed per guideline indications, and fewer deaths were noted in the postintervention period than in the pre-intervention period, but the differences were not statistically significant (Table 2).
Because the postintervention period had more male patients and older patients, we evaluated the outcomes using a logistic regression model controlling for both age and gender. The odds of surgery or transfer for surgery for patients in the postintervention period were 4.88 (95% CI, 1.20-19.79; P = .027) as compared with the pre-intervention period, and the odds ratio for death among patients in the postintervention period compared with the pre-intervention period was 0.40 (95% CI, 0.09-1.69; P = .21) (Figure 2).
