Figure 1.
Flow chart of the two patient groups for establishment of a spontaneous PMVSD closure prediction scoring system.
(VSD, ventricular septal defect; CHF, congestive heart failure).
Table 1.
Number of patients with different follow up periods in derivation and validation cohort.
Table 2.
Baseline characteristics (categorical variables) of 1873 patients in the derivation cohort.
Table 3.
Baseline characteristics (continuous variables) of 1873 patients in the derivation cohort.
Table 4.
Results of multivariate Cox regression analysis in the derivation cohort.
Figure 2.
Plots show SC probabilities occurring within 1 year(□) and 3 years(○) among PMVSD patients in the derivative cohort, plotted against the scoring system.
The LOESS fit lines (the solid line for 1-year and the dashed line for 3-year) using 50% fit plots show the trend of SC probability against the score. (SC, spontaneous closure).
Figure 3.
Comparison among PMVSD patients of high probability(△), intermediate probably(○) and low probability(□) occurring SC by Kaplan-Meier's method.
A: Probability of defect remaining open in the deviation cohort (p<0.001). B: Probability of defect remaining open in the validation cohort (p<0.001).
Table 5.
Scoring system for prediction of spontaneous PMVSD closure.
Table 6.
Probability of SC associated with each score.
Table 7.
Estimated probabilities and observed SC rates in derivation and validation cohort according to score groups.