Pericardial effusion (PE) prognosis depends on the underlying etiology. We sought to assess the outcome of patients with chronic (>3 months), large (diastolic echo-free space >2 cm), idiopathic (without apparent etiology), C-reactive protein (CRP) negative PE, a topic where data are lacking. A total of n = 74 patients were enrolled in this study. Patients were treated by pericardiocentesis (n = 39) or surgical pericardial "window" (PW) (n = 13) or conservatively (n = 22). The median follow-up was 24 months (interquartile range: 15 to 38). Among those patients who had PE drained (n = 52), PE re-accumulation occurred in 32 cases (61.5%) and the rate was significantly higher in the pericardiocentesis subgroup (76.9% for pericardiocentesis vs 15.4% for PW group, p <0.001). Patients with re-accumulation had longer disease duration (32.1 ± 25.7 months vs 19.5 ± 23.8 months, p = 0.01), higher maximum PE diameter (32.2 ± 9.4 mm vs 26.1 ± 4.9 mm, p = 0.003) and larger PE volume drained at baseline (1,912 ± 707 mL vs 1,508 ± 387 mL, p = 0.04). Large PE re-accumulation occurred in 41% of patients who underwent pericardiocentesis and in 7.7% of those who underwent PW. In Cox survival analysis the only independent predictor of fluid re-accumulation was the type of intervention, with PW being associated with significantly reduced risk (hazard ratio 0.115, 95% confidence interval 0.015 to 0.875, p = 0.037). Major complications needing treatment were recorded in 12.8% and 15.4% (p = 0.999) of patients who underwent pericardiocentesis and PW, respectively. Moreover, invasive procedures were not helpful in establishing new diagnoses and guide treatment. In conclusion, in asymptomatic patients with chronic, large, hemodynamically insignificant, CRP negative, idiopathic PE, conservative management seems a more reasonable approach in most cases.