Among 1122 patients (median age, 57 years; interquartile range, 51-63 years; 827 women), 330 (29.4%) had simple and 792 (70.6%) had complex nodules; 496 (44.2%) with mutated EGFR (426 with complex SSNs); and 78 (7.0%) patients experienced death or lung cancer-related recurrence or metastases (77 with complex SSNs). The updated classification showed an independent correlation with both the prognosis of patients (hazard ratio [HR], 9.841; 95% confidence interval [CI]: 1.284, 75.405; p = 0.028) and the EGFR mutation status of SSNs (odds ratio, 1.68; 95% CI: 1.08, 2.59; p = 0.020). The multivariate logistic model outperformed the updated classification in predicting EGFR mutation status, with AUCs of 0.740 (95% CI: 0.699, 0.780) and 0.618 (95% CI: 0.583, 0.653), respectively. Patients with complex SSNs had a significantly lower 5-year DFS rate than those with simple SSNs (HR, 32.93; 95% CI: 4.58, 236.80; p < 0.001). Time-dependent ROC analysis showed that the previous classification of SSNs was consistently superior to the updated classification in predicting the prognosis of patients throughout the follow-up period, with maximum AUCs of 0.723 and 0.657, respectively. Additionally, the multivariate Cox regression model showed the highest prognostic association with SSNs and optimal calibration over time, with a maximum time-dependent AUC (iAUC) of 0.850.