Baseline Characteristics of Selected Participants
A total of 514 subjects met the inclusion criteria (see Figure 1 for a flow chart). Table 1 is the baseline characters of the participants which is based on different groups of the systolic blood pressure and diastolic blood pressure. The average age of the study subjects is 61.92 ± 9.49 years old, of which 67.12% are male. 94 subjects had a MACE event within 3 years, and the occurrence rate was 18.29%. 9 stent restenosis, 3 cardiogenic death, 4 acute myocardial infarction, 15 revascularization, 5 stroke, 8 heart failure, and 90 recurrent chest pain. The incidence of MACE in the low, medium, and high systolic blood pressure groups were 15.79%, 19.88% and 19.19% respectively. The incidence of MACE in the low, medium, and high diastolic blood pressure groups were 16.96%, 17.06% and 20.81% respectively. No statistically significant differences were detected in Alcohol consumption, Smoke, AF, DM, heart failure history, Myocardial bridge, number of stents, Cl-, creatinine, FT3, FT4, HDL-C, LDL-C, Lipoprotein, K+, LVEF, TC, TG, UA, Urea, ACEI, Aspirin, Nitrates among different SBP groups (P >0.05). The group with the highest systolic blood pressure had the oldest average age, the highest incidence of myocardial infarction, the highest BMI, Cysteine protease inhibitor C, TSH, and Na+ values, but the lowest incidence of cerebral infarction. No statistically significant differences were detected in Sex, Alcohol consumption, Smoke, Degree of education, AF, DM, heart failure history, Myocardial bridge, number of stents, Cl-, creatinine, Cysteine protease inhibitor C, FTH, HDL-C, LDL-C, Lipoprotein, K+, LVEF, Na+, TC, TG, Urea, ACEI, Aspirin, b-blocker, Clopidogrel, Nitrates, Ticagrelor among different DBP groups (P >0.05). The group with the highest diastolic blood pressure had the lowest average age, the highest incidence of myocardial infarction, and the highest BMI, TSH, and UA values.
Result of Univariate Analysis
We listed the results of univariate analyses in Table 2. By univariate binary logistic regression,
we found that age (1.03, 1.01-1.05), DM (2.01, 1.82-2.20), Cerebral infarction (1.85, 1.62-2.08) were positively correlated with the occurrence of MACE. LVEF (0.97, 0.95-0.99) was negatively associated with the occurrence of MACE.
Results of Unadjusted and Adjusted Binary Logistic Regression
In this study, we constructed three models to verify the relationship between systolic or diastolic blood pressure and the occurrence of MACE events respectively. The specific values of the effect size and 95% confidence interval are shown in Table 3 and Table 4. In the crude model, the model-based effect size can be interpreted as a change in the risk of a MACE event for every unit change in blood pressure. For systolic blood pressure, in the crude model and model 2, for 1 mmHg increase in systolic blood pressure, the risk of MACE events increases by 1%(1.01, 95% CI 1.00-1.03). In model 3, for 1 mmHg increase in systolic blood pressure, the risk of MACE events increases by 6% (1.06, 95% CI 0.99-1.14). For diastolic blood pressure, in the crude model, the occurrence of MACE events does not change with changes in diastolic blood pressure (1.00, 95% CI 0.98-1.03). In model 2, for 1 mmHg increase in diastolic blood pressure, the risk of MACE events increases by 1% (1.01, 95% CI 0.98-1.03). In model 3, for 1 mmHg increase in diastolic blood pressure, the risk of MACE events is reduced by 2% (0.98, 95%CI 0.90-1.06). We converted the blood pressure value into a categorical variable (Tertile of SBP and DBP) for the purpose of sensitivity analysis. We performed a trend test on the P value, and the results were consistent with the result when blood pressure as a continuous variable.
Results of Nonlinearity of Blood Pressure and MACE
In this study, we analyzed the non-linear relationship between the blood pressures of patients after PCI and the occurrence of MACE events within 3 years (Figure 2 is systolic blood pressure, Figure 3 is diastolic blood pressure). Smooth curve and the result of the Cox proportional hazards regression model with cubic spline functions showed that the relationship between systolic blood pressure and MACE was nonlinear after adjusting for sex, smoking history, drinking history, education level, atrial fibrillation, diabetes, heart failure, myocardial infarction, myocardial bridge, cerebral infarction, number of stents implanted, medication after discharge, BMI, Cl-, Creatinine, Cysteine protease inhibitor C, FT3, FT4, TSH, HDL-C, LDL-C, Lipoprotein, K+, LVEF, Na+, TC, TG, UA and Urea. We used both binary logistic regression and two-piecewise binary logistic regression to fit the association and select the best fit model based on P for the log likelihood ratio test.
When we analyzed the relationship between diastole and MACE events, the log-likelihood ratio test P was greater than 0.05. This shows that there is no significant non-linear or linear relationship between diastolic blood pressure and MACE events. When analyzing the relationship between systolic blood pressure and MACE events, because the P of the log-likelihood ratio test is less than 0.05, we chose a two-part binary logistic regression to fit the relationship between systolic blood pressure and the occurrence of MACE events in 3 years. Through the two-part binary logistic regression and recursive algorithm, we calculated the inflection point was 134. On the left side of the inflection point, the effect size and 95% CI are 1.22 and 1.04-1.43, respectively (P=0.017). The impact size and 95% CI at the right inflection point were 0.96 and 0.83-1.11 (P=0.604), respectively (Table 5).